Claims of sexual injury are always viewed through the twin lenses of sexual morality and sexual politics. Central to each is a narrative of what constitutes a sexual harm.

Traditional society assigned to women the unenviable role of policing sexuality. To this end, its legal system assessed sexual harm from the perspective of the double standard, which admired men and stigmatized women for engaging readily in sex. Sexual harm was thus a wrong against a woman’s chastity and rape was the only recognized sexual injury. A rape conviction required showing not that a man’s motives were blameworthy but that a woman’s virtue was beyond reproach, her lack of consent demonstrated by her sexual history and by a display of the “utmost” resistance.1

First-wave feminists of the Victorian and Progressive eras rejected the double standard and its inevitable division of women into two classes, “the protected and refined ladies and those poor outcast daughters of the people whom [men] purchase with money.”2 While accepting the notion of sexual harm as an injury to chastity, they advocated a sex-neutral standard of sexual restraint grounded in Christian doctrine. The next wave of sex law reform began during the 1950s and was grounded in a sex-neutral standard of sexual autonomy that freed women from the constraints of the double standard but failed to provide a compelling secular narrative of sexual harm to replace religious doctrines justifying sexual restraint.3

First-wave feminism and sexual autonomy theory achieved some important sex law reforms, notably the raising of the age of consent from its appalling historical norm of ten or twelve. Still, the law of rape, with its focus on the victim’s behavior, proved hard to change. Courts began in theory to recognize civil actions for sexual harassment, but so grudgingly that recovery was virtually impossible.

In response to these failures, feminists in the seventies argued that sex-neutral theories of sexual autonomy failed to identify the critical role of sexuality in the subjugation of women. Male sexual advances, they argued, were often or even always motivated not by sexual passion but by a desire to humiliate and subordinate women. This dominance framework supplied a secular theory of sexual harm, an element critically missing from early discussions of sexual autonomy. The source of sexual harm, from this perspective, was located in the intent of the male actor, shifting inquiry from female to male motives.

The prevailing public narrative of sexual harm became an odd synthesis of themes from dominance theory and sexual autonomy theory. That narrative accepted one key principle from the most radical dominance model, the strong misogyny narrative: all sexual harm resulted from men’s generalized desire to degrade and exert power over all women. However, public opinion rejected the idea that such motives pervaded heterosexual interaction. Thus, men were divided into two groups: good actors conformed to the norms of the sexual autonomy model, while bad actors fit the model of strong misogyny theory and were driven by an all-encompassing animus towards women as a group.

By providing a model of sexual harm, this hybrid misogyny model succeeded where autonomy theories had failed and produced a seismic shift in both public and judicial attitudes. Under its influence, the law of rape finally began to undergo a period of significant reform, with changes such as rape shield laws that shifted legal inquiry away from the character of the victim. Perhaps most dramatically, courts began to increase substantially the scope of liability for sexual harassment, authorizing recovery for non-economic harm through hostile environment theory.

However compelling, I argue, the hybrid misogyny narrative is incomplete. By locating the problem of sexual harm solely in the actions of deviant misogynists, it impeded recognition of the damage that can be done by flawed but not evil men, especially in situations of power created by the workplace. When applied to workplace settings,4 the hybrid misogyny narrative paved the way to judicial expansion of sexual harassment liability, but also to some deeply misguided doctrines.

In this article I will discuss three of these distortions. First, the focus on misogynistic motive led courts to overlook the importance of objective power differentials and thus failed to distinguish adequately between supervisor and coworker cases. Early sexual harassment cases imposed liability only in cases of quid pro quo, where benefits or detriments of the job were explicitly conditioned on submission to sexual demands. In extending this doctrine, the Supreme Court might well have preserved a distinction between the acts of supervisors and those of coworkers, but it did not, instead providing a single doctrine, hostile environment theory, that was applicable to both supervisors and coworkers.5 Judicial deprecation of the importance of implied coercion worsened when the Court handed down rules governing agency liability that made recovery for supervisory conduct only slightly easier than for coworker conduct.

Second, the notion that all harm results from misogynistic animus has prevented recovery in cases in which the defendant’s actions, however harmful, were motivated not by animus but by a genuine romantic interest in the plaintiff that was ultimately rejected, sometimes after a consensual affair. At that point, the defendant began to engage in workplace behavior that was harmful to the plaintiff, but that was not, viewed in isolation from the sexual rejection, hostile or misogynistic. Courts typically reject liability in these cases, reasoning that thwarted affection rather than generalized animus towards women motivated the defendant. Yet Title VII does not itself contain any requirement that the plaintiff prove animus but rests liability on a showing that a defendant acted “because of” sex.6 The strong misogyny narrative in effect adds to the plaintiff’s burden by requiring her to prove an element that is not to be found in Title VII or Supreme Court opinions.

Finally, the picture of a unitary sexualized misogyny, directed against all women, has obstructed efforts to develop defensible doctrines governing admission of “me-too” evidence from women other than the plaintiff, which exponentially improve the chances of success by a harassment plaintiff. The law disfavors but does not wholly disallow evidence of prior bad acts, and such evidence has been increasingly allowed in the past ten years under a variety of theories. In response, defendants have recently begun to produce rebuttal witnesses who testify to the defendant’s respectful treatment of women. A strong misogyny theory that treats sexual harassment merely as a specific manifestation of more generalized misogyny cannot justify the exclusion of defendant me-too evidence, since it has tremendous difficulty explaining why a man would single out some women while leaving others alone.

As a purely legal matter, these issues might be handled by carefully targeted doctrinal arguments. Even judges, however, are influenced by narratives. Moreover, public reaction to the #MeToo movement suggests that the misogyny model may be engendering a worrisome backlash. Some of this may be dismissed as the inevitable opposition that reform movements always encounter, but some concerns seem legitimate. The double standard and traditional legal doctrine focused almost entirely on the blameworthiness of the victim’s conduct. Misogyny theory turns the tables too completely, assimilating all sexual harm to the paradigm of rape, a crime of the most profound blameworthiness. Though it effectively blocks attacks on victims, it encourages a moralizing stance towards the conduct of even those defendants whose behavior, though an appropriate basis for legal liability, falls far short of rape. And moral opprobrium in excess of what is warranted erodes public support and invites resistance.

The history of the law of sexual harm suggests that narratives matter, and new narratives are needed. In highly charged areas of law, judges will have difficulty applying doctrines without support from a moral framework that makes sense to them and fits the facts of the case before them. If #MeToo is to usher in a new phase of sex law reform, it must construct new and more nuanced narratives of sexual harm that go beyond misogyny and sexual autonomy.

One central priority is the development of a narrative about what situations are sufficiently coercive to require state supervision through legal intervention. Past waves of legal reform have had great success in passing laws that single out statuses and contexts that are unacceptably coercive. Liberal theorists have labored mightily to provide a more cohesive account of coercion without complete success.7 If the ball cannot be moved on theory, perhaps at least more compelling stories can be constructed.

A second priority is to resurrect what the Victorians knew: most misogynists construct two kinds of girls, the pure and the sullied. Misogyny is not simply about hating women, it is about dividing them.

Finally, and perhaps most importantly, the hybrid misogyny narrative does not advance the extraordinarily challenging cause of providing a secular morality that acknowledges the emotional dangerousness of sex not only when it is casual but even more so when it is not. Sex has consequences, both for men and for women, and a society that fails to provide a moral framework for sexual behavior does so at its peril.

I. Narratives Matter

Title VII prohibits employers from adversely affecting the terms, conditions, or privileges of employment because of sex.8 A sexual harassment complainant must allege and prove behavior that caused harm to the terms or conditions of employment; was attributable to the employer; and had a causal nexus to her sex. In practice, the theory of harm has proven to be the most important of these elements and exerts a kind of halo effect on the theoretically distinct elements of causation, agency, and proof. Courts that find harm seem willing to stretch to find other elements satisfied. Courts that find no harm seem to set impossibly high standards for other elements.

Sexual harm is unavoidably viewed against the background of views about sexuality and relationships between men and women. Over the last two centuries a series of legal efforts have attempted to control sexual harms,9 each responding to the dominant moral framework of its time with a narrative on the nature and causes of these harms.

The overwhelming majority of human societies have imposed a double standard on the sexual behavior of males and females, with males invariably the beneficiary of the more permissive norm.10 Nonetheless, the degree of inequity as well as numerous other details of the double standard has varied widely among societies. For present purposes, the relevant double standard is that of the Anglo-American world prior to the mid-twentieth century, to which I refer as traditional, although it is not in all respects a universal tradition.

Traditional Anglo-American sexual mores reflect a tension between two perspectives, the Christian and that of popular mores. Both were united in assigning importance to the regulation of sexual behavior but they differed in their demands on men’s self-control and in turn in the burdens they placed on women. Christianity rejected the double standard, preaching chastity, in the sense of abstinence from non-marital sex, for both men and women. Social norms likewise restricted female sexuality to marriage but leant strongly towards the double standard. Men were expected, permitted, and encouraged to pursue with vigor every opportunity for sexual activity, while women were correspondingly expected to resist these efforts.11 In the public’s mind, this double standard was simply an entitlement of masculinity,12 while intellectual and religious perspectives generally saw sexual restraint as a requisite of family stability.13

The balance between Christianity and the double standard vacillated over the years, but before the Victorian era the double standard dominated both public opinion and the law. Males were assigned virtually no responsibility for controlling their own sexuality, and social norms accepted the existence of a class of vilified prostitutes14 to satisfy male lust in order to protect the virtue of good women. The burden was placed on individual women to demonstrate that they belonged to the protected class of ladies rather than among the fallen.

Sex law reflected the tensions between these frameworks. Under the influence of Christianity, traditional Anglo-American law nominally placed stringent limits on sexual activity by both sexes, prohibiting fornication, adultery, bigamy, and contraception. However, these laws operated more harshly against women than men both by their terms and by custom.15 Other laws forthrightly buttressed the double standard and protected the right of males, especially wealthy ones, to wide sexual access to females. Laws against prostitution fell far more harshly on sex workers than on their clients.16 For hundreds of years, the Anglo-American age of consent was ten or twelve.17

The law of rape reflected the wholly male concerns that women were inclined to make false charges of rape that were hard to rebut.18 The core definition of rape was sexual intercourse that was both forcible and without consent.19 Victims were, as is well known, required to resist strenuously, often to the utmost, a condition not always met even in instances of submission to credible threats of deadly force.20 Many jurisdictions imposed further requirements unique to rape cases: immediate complaints; eyewitnesses or physical evidence; and cautionary instructions to the jury.21 The victim’s prior behavior, both sexual and otherwise, was open to virtually every possible type of prejudicial questioning at trial.22

The balance began to shift with the Victorians, who stressed the critical role that women played in the social order. The image of the angel in the house23 whose domestic virtues civilized men24 has been much mocked by later feminists as a cult of domesticity that fetishized female submissiveness. Compared with what came decades later, these early forms of separate spheres ideology had elements that were cramped and confining. Compared, however, with what came before, its vision of women’s moral superiority was a radical step toward improving the status of women. Nineteenth century reformers such as Frances Willard, Josephine Butler and Jane Addams broadened the notion of woman’s sphere to include public reform efforts, to which Addams referred as “public housekeeping.”25 These reformers argued passionately for a single standard of sexual behavior:

[N]umbers even of moral and religious people have permitted themselves to accept and condone in man what is fiercely condemned in woman. And do you see the logical necessity in this? It is that a large section of female society has to be told off—set aside, so to speak, to minister to the irregularities of the excusable man. That section is doomed to death, hurled to despair; while another section of womanhood is kept strictly and almost forcibly guarded in domestic purity. [P]ublic opinion [must], both in theory and practice, shall recognize the fundamental truth that the essence of right and wrong is in no way dependent upon sex, and shall demand of men precisely the same chastity as it demands of women.26

These thinkers were the first to see the link between sexual and economic oppression. Women’s rights activists condemned wealthy men who regarded access to working women as a class privilege. The continuum between the exploitation intrinsic to prostitution and the sexual exploitation of women outside sex work was widely noted and understood as the result of an economic system that denied women access to economic opportunities.27

A movement to reform sex law was an important part of the agenda of the first-wave women’s movement. The durability of the traditional doctrine of consent made the generally applicable law of rape impervious to these Victorian efforts: if consent could be contested at trial, prosecution was seldom possible. Reformers therefore targeted a narrower set of cases, combining their larger narrative attacking sexual exploitation with carefully crafted legal arguments applicable to particular contexts in which consent could be said to be wanting as a matter of law.

A central achievement was the expansion of the law of statutory rape.28 Between 1885 and 1920, all US states raised the age of consent from between seven and twelve to between sixteen and eighteen.29 Though situating this campaign in the context of larger moral issues, reformers repeatedly stressed a central inconsistency in the legal system’s treatment of youth and incapacity. Boys and young men were protected until age twenty-one from an imprudent decision to enter even a trivial contract, while a girl over the age of ten who had even coerced sex was forced to bear life-altering consequences, such as pregnancy and unmarriagability, without legal protection.30

A second and more complex set of legal protections consisted of civil and criminal actions for seduction. The early civil actions were compromised by patriarchal attitudes, since only the victim’s male guardian had standing to sue, but over time both the defensibility and the efficacy of these laws increased as women obtained the right to represent themselves.31 Seduction theories were somewhat varied in their requirements, but many were dependent on the existence of misrepresentation. Thus, like statutory rape law, they could be defended as consistent with the rules governing non-sexual offenses. Relatively few doctrines specifically addressed the coercive nature of the workplace, but most in practice developed into a tool that protected young working women, and one interesting Missouri statute criminalized sexual relations between employers and young women employed in domestic service.32 Women’s rights groups pressed for the passage of criminal seduction laws,33 which were widespread by the late nineteenth century, though later repealed for supposedly exposing men to female exploitation.34

The last cohort of first-wave feminists in the early twentieth century placed more emphasis than their predecessors on sexual freedom, advocating a range of views from the mildly permissive35 to the radical support of free love.36 After the passage in 1920 of the Nineteenth Amendment, feminism entered a quiescent period, but public sexual norms gradually relaxed, in part because of improvements in the quality and availability of reliable contraception.37 A wide variety of thinkers including academics like Alfred Kinsey38 and Wilhelm Reich,39 the writers of marriage manuals,40 and public figures like Hugh Hefner41 and Helen Gurley Brown42 advocated increasingly permissive norms of sexual behavior. Like their Victorian-progressive predecessors, they endorsed a sex-neutral standard. However, rather than extend the obligation of chastity to men, they hoped to provide women with the sexual freedoms traditionally reserved to males. Their perspective, now referred to as the sexual autonomy model, made fundamental the right of all individuals to decide for themselves with whom and under what circumstances to have sex.

In the late 1950s, a new wave of sex law reformers began their work, this time based on the norm of sexual autonomy. The expansion of autonomy-based rights began with reforms in the law of rape43 and led to groundbreaking protections of women’s right to sexual pleasure, notably in cases Constitutionally protecting access to contraception and abortion.44 But the liberal alternative to religious sexual morality gave primacy to freedom of choice and provided no alternative narrative of sexual harm or voluntary restraint. In consequence, though women gained much needed freedom to engage in desired sex, far less progress was made in protecting them from undesired sex. Without such a narrative, the scales were inevitably tipped against doctrines that could be seen as restricting the freedom to have sex. Some progress was made in the expansion of statutory status-based offenses that involved the abuse of positions of power.45 But the law of rape, with its focus on the victim’s behavior, proved hard to reform, and the sexual double standard continued to infect the application of even revised doctrine.

The first sexual harassment cases were brought in the early 1970s and met a mixed reaction from courts. Virtually all involved the quid pro quo conduct of a supervisor who conditioned the terms of employment on the toleration of sexual advances, often quite degrading in nature. Cases rejecting liability typically expressed the same concern for preserving male sexual opportunities that animated the worst aspects of rape law, and suggested that liability for even quite egregious behavior would foreclose all sexual activity in the workplace.46 These courts seem to have concluded that the defendant caused no harm to the plaintiff beyond the inconvenience that sometimes attended normal heterosexual interaction. This skepticism about the gravity of injury spilled over into unrelated elements of the cause of action: these cases found that plaintiffs had failed to prove not only the elements of harm but causation and agency liability.

Those early courts that accepted sexual harassment as a ground of recovery appear to have done so from a different perspective on the calculus of harm. They did not assess the sexual harm to women as more serious but rather saw the problem of economic discrimination as of greater importance. Their conception of injury evidently colored their analysis of the superficially unrelated requirements of causation and agency, which they were more likely to find satisfied.47 But even these courts imposed a requirement of tangible economic harm or at least the explicit threat of such harm. Abusive sexual behavior by supervisors without explicit threat or actual retaliation was without remedy. Sexual autonomy theory could, on a purely theoretical level and with some struggle, justify legal recognition of highly limited claims of sexual harassment, but it could not help judges see cases through the eyes of victims. Its sex-neutral picture of equal sexual agency focused primary on increasing freedom, leaving too many wondering why the work environment transformed sexual behavior that was acceptable in most spheres into a legal harm.

The watershed change in the law of sexual harm began in 1975. In Against Our Will, Susan Brownmiller argued that men did not rape because they wanted sex, but rather used rape as a means to humiliate and subordinate women.48 Indeed, much or even all of what passed for consensual sex was in fact subtly disguised rape, a tool of oppression rather than an expression of desire. The new paradigm in rape law laid the ground work for Catherine MacKinnon to refocus the sexual harassment debate from formal equality to the power structures underlying sex at work. Her 1979 landmark book, Sexual Harassment of Working Women, had been influential in draft form even before its publication. Her theory had two distinct components. The first concerned the victim’s experience: the existence of the workplace power relationship could make otherwise non-problematic sexual behavior coercive.49 At the same time she began to advance the thesis that Brownmiller had developed in her work on criminal rape: Men did not seek sex and incidentally dominate women. They sought sex in order to dominate women.50 The inherently coercive nature of the workplace was simply a useful tool to effectuate the underlying goal of dominance. In her early work, MacKinnon focused more or less equally on the harasser’s impulse to subordinate and the role of the coercive environment in the experience of the powerless. In her later work, her focus shifted to the desire to dominate, and the institutional setting became unimportant:

The uncoerced context for sexual expression becomes as elusive as the physical acts [of sexuality and violence] come to feel indistinguishable. [R]ape is defined as distinct from intercourse, while for women it is difficult to distinguish the two under conditions of male dominance.51

From this perspective, the coercive nature of the workplace was almost irrelevant, since harm resulted from the fact that men’s motives were so pernicious and their power so omnipresent.

Some feminist writers on sexual harassment accepted only elements of the dominance model, notably the view that sexual harm is motivated by misogyny. However, the greatest influence on the public narrative came from MacKinnon and her sometime co-author Andrea Dworkin. MacKinnon and Dworkin did not stop simply at the observation that sexual harm can be the result of a defendant’s urge to oppress, but argued instead that it must be the source of all sexual harm. Moreover, the desire to subordinate was inevitably a generalized desire to degrade and humiliate all women and such motives were pervasive in heterosexual relationships.52 I will call the view that both of these hypotheses are true the strong misogyny narrative of sexual harm.

Dominance theories captured a fundamental truth: sexual harm can result from sexual advances motivated not by erotic desire but by dominance and misogyny. The dominance model received widespread and generally respectful public attention. The idea that dominance rather than desire motivated sexual advances in the workplace was quickly accepted by observers across the political spectrum.53 It thus supplied the secular theory of harm that sexual autonomy models had failed to provide. This new narrative provided public discussion with a powerful way of rethinking sexual misconduct that was acceptable even to those who did wholly abandon the double standard. The double standard absolved men from any responsibility to control their sexual impulses—that task was left to women, who were assigned the unappealing role of sex police. But every society expects men as well as women to control their violent impulses, and if rape was a crime of violence its perpetrator had breached the most central of societal norms. Similarly, society expects its members to keep their nonviolent but aggressive impulses in check, so that sexual behavior short of rape is more easily seen as misconduct if motivated by hostility rather than erotic passion.

The focus on motive also placed the issue of consent in a new light. Men might argue that women desired (and consented to) sex more often than they would admit, but few were willing to claim publicly that women desired victimization. If sexual harms were the result of misogyny, the questions of consent and unwelcomeness receded in importance, since the wrong inhered in the accused’s intent rather than the victim’s failure to resist. By persuading the public that at least some rape cases were motivated by hatred rather than lust, the dominance thesis paved the way for crucial rape law reforms like victim shield statutes. Similarly, a focus on animus in workplace harassment blunted the impulse to ask whether the victim had encouraged the behavior at issue.

The dominance theory in general and MacKinnon’s work in particular received widespread and generally supportive coverage in the popular press. At the same time, even sympathetic observers seldom accepted MacKinnon’s general view of relations between men and women.54 In the words of one, “what McKinnon represents is the embattled, bleak, martyred side of feminism. [H]er view is so narrow and her attitude so wound-licking that we tend to get awfully weary of her version of “unmodified feminism” early on.”55 Her dark picture of sex between men and women was often criticized as a step backwards for women’s efforts to achieve sexual pleasure on the same terms as men.56 Public opinion thus accepted a circumscribed version of dominance theory, which was understood to describe the conduct and motives of a subset of men, and this picture was helicopter dropped into the otherwise prevailing model of sexual autonomy. The next section examines how this ambivalent acceptance of the misogyny model played out in the case law.

II. Supervisors, Coercion and Hostility: The Wrong Turn

The idea that sexual misconduct is about power rather than sex is compelling and easily understood, and it is now a commonplace of public discussion. This simple thesis did nothing short of revolutionize sex law. But its simplicity obscured several very real complexities, such as the possibility of harm without misogyny and the relevance of coercion to alternative conceptions of harm. To further complicate matters, few academics, much less the courts and the public, accept the strong misogyny model in its entirety, and the sexual autonomy model remains important. Examining the evolution of the core doctrines of sexual harassment reveals how the new narrative developed an awkward combination of autonomy and misogyny principles. The resulting hybrid misogyny model recognized sexual harassment in theory while providing little relief in practice.

By the end of the 1970s, courts had come to accept as a form of sex discrimination the explicit conditioning of job benefits on toleration of sexual conduct. Title VII’s prohibition of this scenario was put on a sound doctrinal footing by Barnes v. Costle,57 which stressed the critical role of but-for causation. Discrimination occurred because the plaintiff would not have suffered the disadvantageous outcome but for her sex.58

After Barnes, the next critical question was the proper treatment of cases without an explicit quid pro quo that caused tangible economic injury. In Sexual Harassment of Working Women, MacKinnon had proposed a model of non-quid pro quo cases59 which she later called environmental harassment and described as “sexual insult and aggression.”60 MacKinnon did not entirely ignore the issue of coercion in environmental discrimination, but she identified as the coercive element not the supervisory relationship, but women’s generally poor labor market prospects.61 In 1980,62 the EEOC issued Guidelines that followed MacKinnon’s distinction, defining sexual harassment as unwelcome sexual conduct that either contained a quid pro quo, whether implicit or explicit,63 or that had “the purpose or effect of unreasonably interfering with an individual’s work performance or creating an intimidating, hostile, or offensive working environment.”64

A number of lower courts soon adopted hostile-environment theory, and the Supreme Court endorsed it in the 1986 case Meritor Savings Bank v. Vinson,65 in which the plaintiff’s legal team included MacKinnon.66 The plaintiff, Mechelle Vinson, testified that she had been intimidated into a sexual relationship with her supervisor, Sidney Taylor, that had included rape. However, there had been no concrete retaliation or explicit threat of such,67 and the defendant argued that lack of economic injury precluded liability.

The central holding of Meritor is that discrimination with respect to the terms and privileges of employment includes not only “tangible loss” of “economic character,” but also “psychological aspects of workplace environment.” The Court’s opinion, however, goes further than this by following the plaintiff’s brief, adopting the misogyny model’s emphasis on motive, and changing the narrative of workplace sex. Where unsympathetic courts had seen sex, the Meritor Court saw sexual abuse: “Surely, a requirement that a man or woman run a gauntlet of sexual abuse in return for the privilege of being allowed to work and make a living can be as demeaning and disconcerting as the harshest of racial epithets.”68 If defendants were motivated by the desire to abuse rather than the desire for sex, their needs no longer deserved weight in the social calculus, and thus the scope of liability could be enlarged.

Meritor and subsequent cases reflected the emerging understanding that sexual advances can be sex discrimination, but they did so by a route that embodies several troublesome premises. First, they suggested that an abusive motive is a key element of the cause of action, in stark contrast to non-sexual Title VII doctrine, in which only but-for causation rather than hostility is required. Second, the animus-based theory of harm seriously deprecates the importance, from the employee’s perspective, of supervisory authority. Quid pro quo doctrine is based on the significance of the power relationship, and the Meritor Court might have chosen to expand existing quid pro quo doctrine, as the EEOC Guidelines suggested, to acknowledge the implicitly coercive nature of any supervisory relationship.69 Vinson testified that she reluctantly acquiesced to Taylor’s demands and did not report him for fear of losing her job: had Taylor been a coworker instead of a supervisor, Vinson would have been far more likely to resist. Instead the Court chose to apply hostile environment doctrine, which focuses not on the context but on “hostile” motivation of the harasser.

The Court solidified the doctrinal emphasis on the defendant’s motivation in Harris v. Forklift, which held that to be actionable under hostile environment theory, an environment must be one “that a reasonable person would find hostile or abusive.”70 The court imposed a totality of the circumstances test whose four factors made no reference, direct or indirect, to the existence of supervisory authority.71 Hostile work environment theory thus applied in an identical fashion to the behavior of coworker and supervisors, and the special role of the supervisory relationship was limited to cases of explicit quid pro quo. This was a striking shift, since even earlier courts that rejected liability were remarkably candid in noting the coercive nature of the supervisory relationship.72

This sole focus on the motives of the harasser, to the exclusion of circumstantial factors creating coercion, might have had done little harm to future plaintiffs had the Court accepted the strong misogyny model’s view that malign motives were pervasive. But the Court, like most Americans, rejected this view and repeatedly indicated that it did not intend to provide recovery for all sexualized behavior73 or even some mildly offensive behavior.74 In effect the Court adopted a sexual autonomy model that permitted adults to engage in sexually tinged conduct as long as that behavior was kept within an acceptable range.

This doctrinal narrative, which probably tracked public sentiment as well, might thus be called a hybrid misogyny model. It took one key principle from the strong misogyny narrative: all sexual harm resulted from the desire to degrade and exert power over all women. At the same time, it assumed that respect was the norm in sexual relations between men and women. Thus, the judicial narrative distinguished two types of men. Good men, the majority, conformed to the norms of the sexual autonomy model and its consent-based morality. Only a relatively small group of bad men were motivated by animosity towards women, and it was this group only with whom the law was concerned. The malign motives of these bad actors were the source of sexual harm, without regard to contextual factors creating coercion.

The problems of the hybrid model can be seen when the current standard is applied to specific facts. In Baskerville v. Culligan Int’l Co.,75 the plaintiff, Valerie Baskerville, had been subjected to a constant stream of sexual speech and indications of sexual interest by her supervisor, Michael Hall.76 For example, at one point, he told her that “his wife had told him he had ‘better clean up my act’ and ‘better think of you as Ms. Anita Hill.’”77 On another occasion, the announcement “May I have your attention, please” was broadcast over the public address system. Hall stopped at Baskerville’s desk and said, “You know what that means, don’t you? All pretty girls run around naked.”78

The Seventh Circuit, in an opinion by Judge Posner, took the unusual step of overturning a jury verdict for plaintiff.79 It reasoned that the defendant was “not a sexual harasser” but merely “not a man of refinement” and “a man whose sense of humor took final shape in adolescence.”80 Moreover, Posner stated, “[t]he comment about Anita Hill was the opposite of solicitation, the implication being that he would get into trouble if he didn’t keep his distance.”81 Noting that context might change the effect of the remarks, the opinion nonetheless concludes “there is no suggestion of any other contextual feature of their conversations that might make [the defendant] a harasser.”82 Commenting that Hall “never said anything to her that could not be repeated on primetime television,”83 Posner concluded, “only a woman of Victorian delicacy—a woman mysteriously aloof from contemporary American popular culture in all its sex-saturated vulgarity—would find [the defendant’s] patter substantially distressing.”84

In some sense, Posner was correct. The defendant was not clearly a misogynist and quite possibly simply an immature and silly man. But this fact does not deserve the importance he gives it unless existing hostile environment doctrine adds to Title VII a requirement of animus that has been rejected in other contexts.85 What should have been critical was whether the conduct would not have been directed to her but-for her sex (and it clearly would not have been) and whether she suffered harm sufficient to trigger Title VII. The latter question can only be answered the context of the fact that Hall was Baskerville’s supervisor and that the comments were explicitly directed to her—she was not watching a TV show. Most women, I will venture, find the behavior of Michael Scott in The Office hilarious rather than offensive, in part because the show mocks rather than condones his behavior.86 The same behavior by an actual supervisor would provoke a very different reaction. Hall in effect told Baskerville that he might have propositioned her had it not been for his wife’s warning, a comment that would have been unsettling even in a social setting and was downright scary coming from a supervisor. Posner argued that Hall’s implication that he would not harass Baskerville eliminated any sexual threat from the situation. How far would Posner take this reasoning: would he be similarly dismissive of the statement “I’ve fantasized about forcing myself on you, but don’t worry, I won’t”? The opinion is all the more remarkable because Posner has elsewhere shown great insight into the humiliating nature of similar interactions.87

Baskerville illustrates the pitfalls of emphasizing motive but might be dismissed as a singular opinion. The next Section considers three areas in which the problem of the strong misogyny theory have or are threatening to create broader doctrinal problems.

III. Three Doctrinal Problems

A. Agency Liability

The Supreme Court missed another opportunity to draw a bright line between supervisor and coworker harassment when it considered the rules governing employer liability. Greater liability for supervisor than coworker conduct would not have erased the problems caused by Meritor, but it would have focused employer attention on the main problem. Instead, the court chose blur further the boundaries between supervisory and coworker conduct in the twin 1998 cases of Burlington Industries, Inc. v. Ellerth88 (Kennedy) and Faragher v. City of Boca Raton89 (Souter).

Justice Kennedy’s Burlington opinion began well, rejecting the use of the categories quid pro quo and hostile work environment in determining vicarious liability, holding that courts should instead look to agency law and the purpose of Title VII.90 Agency law, as summarized in the Restatement (Second) of Agency §219, provides strict employer liability for the acts of employees “committed while acting in the scope of their employment.”91 Although the question was somewhat closer than the opinion suggested,92 the Court was not clearly wrong in holding that sexual harassment by a supervisor was not conduct within the scope of employment.

The opinion then examined the distinction between coworker and supervisor conduct under the Restatement’s provision for liability when the employee “was aided in accomplishing” the wrongdoing by the existence of agency relation, even where the acts were outside the scope of employment.93 The opinion noted that a generous interpretation of this rule would imply strict liability for all coworker harassment and declined to adopt this view, noting that neither the EEOC nor other courts had advocated this approach.94 The Court acknowledged that supervisors who take tangible employment actions are more clearly aided by the agency relationship than are coworkers: “a supervisor’s power and authority invests his or her harassing conduct with a particular threatening character.”95 The Court declined, however, to find that all harassing supervisors were “aided” by the supervisory relationship on the grounds that “there are acts of harassment a supervisor might commit which might be the same acts a coemployee would commit, and there may be some circumstances where the supervisor’s status makes little difference.”96

The Court concluded that agency law was insufficiently clear to provide a rule governing the proper scope of employer liability, and instead looked to the policies underlying Title VII.97 It stated that the central such policy was conciliation and the avoidance of lawsuits,98 and therefore imposed a kind of negligence standard in cases in which a supervisor had not taken tangible job action.99 Employers would not be liable if (i) they exercised reasonable care to prevent and correct promptly any sexually harassing behavior and (ii) the plaintiff-employee unreasonably failed to take advantage of any preventive mechanisms the employer provided.100

These agency rules have proven nothing short of catastrophic for sexual harassment victims. Survey evidence shows that victims believe that their reports will be at best ignored and will more likely subject them to retaliation.101 One study found that more than 75% of complainants encountered retaliation.102 Victims are caught in a double bind. Even short delays in reporting will be found unreasonable, providing the employer with a complete defense.103 However, employees forfeit Title VII’s provisions against retaliation if they make complaints of sexual harassment before the conduct rises to a level at which it becomes actionable.104 Courts have insisted on prompt reporting of incidents even when the employer has actual notice of egregious public conduct prior to formal reporting.105 Although courts typically impose stringent timeliness requirements on plaintiffs, they are tolerant of significant delays in response by defendants.106 Courts generally reject employee claims that they failed to report because of concerns about futility or retaliation,107 even when these concerns can be substantiated108 or even when a direct threat has been made.109

The Burlington/Faragher rule thus places a burden on sexual harassment plaintiffs utterly unlike any other in Title VII jurisprudence. Title VII’s policy of conciliation is, as a general matter, implemented by requiring plaintiffs to file a complaint with the EEOC and to attempt to reach an administrative settlement before going to court. Burlington/Faragher inexplicably adds a new layer to this process, one which is extremely prejudicial to victims. The employer-procedures defense encourages employers to devise a reporting system that satisfies the courts but discourages complaints.110 It is hard to think of any other area of law in which potential plaintiffs are required to report their concerns and lay out their entire case to a potentially adverse party without the benefit of a neutral intermediary.

At the time of Burlington/Faragher, the Court was far from unsympathetic to sexual harassment complainants. How could it have created such a mess? The misogyny narrative encouraged it to view harassment as an offense of moral turpitude, approaching rape in its seriousness. Had the Court taken this further, adopting a strong misogyny perspective, it would have regarded harassment as part of a larger system of pervasive oppression, and would never have entrusted employers and their agents with the job of protecting women from harassment. But the Court, like the public, rejected the strong dominance theory in favor of a dichotomized picture of a few bad apples in a barrel of good actors. This hybrid view provided no way of thinking about the complex problems that result not from misogyny but from the combination of power imbalance and economic self-interest. Sexual harassment persists because of the misconduct of a few but just as much from the inaction of others. Supervisors are almost by definition more valued by an organization than those they supervise. An employer may fail to act against a supervisor not because it condones his actions but simply because intervention is more costly than looking the other way.

B. Disappointed Affections

Genuine and lasting love can arise at work.111 Public opinion seems solidly supportive of office romance: one survey found that only 4% believe that work relationships are wrong under all circumstances.112 Indeed, between a third and a half of respondents report having had sexual or romantic involvement at work.113 Yet surveys also suggest that attraction and relationships between supervisor and subordinate can be problematic even when genuinely motivated by affection. These relationships account for just under 10% of office relationships,114 are disproportionately dangerous,115 probably account for most of the 6% of workers who have left a job because a romantic relationship with someone at work went sour, and hurt women more than men.116 Public opinion is less approving of relationships between co-workers and subordinates, though only a minority (43%) feel that relationships between supervisors and subordinates are never appropriate.117 These relationships are not uncommon: 22% of workers have dated someone who was their supervisor at the time.118

The perils of romance in the supervisory setting are attested by the significant number of sexual harassment cases involving a defendant whose feelings about the plaintiff seem, on any reasonable interpretation, to have been sincere and respectful romantic interest. In some cases, the plaintiff had initially engaged in a consensual affair. In others the defendant’s interest in the plaintiff was never reciprocated. In either situation, the plaintiff eventually rejected the defendant. At that point, the defendant began to engage in workplace behavior that was harmful to the plaintiff. Sometimes the behavior in these cases is merely wounded— such as avoidance of direct contact that led to less favorable work assignments.119 In other cases the behavior was more antagonistic but would not in itself have risen to the threshold needed for a hostile environment claim.

The adverse consequences of romantic rejection are illustrated by Novak v. Waterfront Comm’n of N.Y. Harbor.120 The plaintiff, Shanti Novak, was a detective with the Waterfront Commission of N.Y. Harbor.121 She became romantically involved with Scott Politano at a time when both held the same rank, in different locations. Eventually the two became live-in partners,122 and during this time, Politano was promoted to Lieutenant and was transferred to Novak’s office, consequently becoming her supervisor.123 Novak terminated her relationship with Politano shortly thereafter.124 After her breakup with Politano, Novak was singled out for unfavorable treatment even after Politano was replaced as Novak’s supervisor.125 Novak was given unfavorable work and shift assignments; was subjected to heightened scrutiny with respect to her work and her requests for overtime pay and sick leave; was the only detective not to receive further formal detective training; was excluded from an email regarding a shooting range schedule; and was the only detective to whom newly hired detectives were not assigned, which was both humiliating and deprived her of the opportunity to learn from the new assignees, who were seasoned detectives from other agencies.126 The situation became worse after she complained to the human resources department.127 Politano refused to communicate with Novak and gave her orders only indirectly through detectives junior to her, and escalated minor work failings into formal written memoranda of counseling.128 Without questioning that Novak was mistreated by Politano or that Politano’s attitude affected the way other supervisors treated Novak, the court concluded that “such mistreatment, while unfair and unfortunate, does not constitute Title VII sex discrimination under existing law.”129

Faced with similar cases, other courts likewise typically reject liability in these cases, reasoning that thwarted affection rather than generalized animus towards women motivated the defendant.130 Yet Title VII does not contain any requirement that plaintiff prove animus, but rests liability on a showing that the defendants acted “because of sex.”131 In principle, the Supreme Court has applied this principle to sexual harassment doctrine: “The critical issue, Title VII’s text indicates, is whether members of one sex are exposed to disadvantageous terms or conditions of employment to which members of the other sex are not exposed.”132 A woman who is denied job opportunities because she has rejected a supervisor romantically is close to the paradigm of quid pro quo harassment. She cannot help but feel pressured to enter or resume a sexual relationship, and such situations clearly pose a serious threat to equal opportunity for women in the workplace. In practice, though, many courts have still viewed these cases through the lens of dominance doctrine, denying recovery on the grounds that the defendant’s conduct was motivated by “personal animosity” rather than sexist animus133 and was thus outside of Title VII’s prohibition on altering the terms and conditions of employment because of sex.134 The misogyny narrative in effect adds to Title VII a requirement that is absent from the statute and Supreme Court opinions.

C. Me-Too Evidence

In the extrajudicial sphere, the #MeToo movement has strikingly demonstrated the power of multiple charges against an individual to succeed where a series of isolated complaints had previously failed. At the same time, a chorus of charges invites a chorus of rebuttals. The defenders of individuals accused of misconduct, such as Brett Kavanaugh and Bill Clinton, often stress evidence that the accused has treated other women well.135 Unfortunately, the strong misogyny theory supports the admissibility of this not-me-too evidence, which is obviously relevant to a charge of generalized animus towards women though less clearly germane to a specific charge of misconduct towards one woman.

This undesirable consequence of generalized misogyny theory has played out in the courts as well. The expression “me-too” evidence predates the #MeToo movement, and refers to evidence of discriminatory behavior, not necessarily sexual, towards an individual not a party to the suit. Its admissibility follows the general rules of evidence: although evidence of prior acts may not be introduced “for the purpose of proving action in conformity therewith,”136 it may be offered for other purposes such as proof of “motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake or accident.”137 The strongest basis for the introduction of me-too evidence is generally thought to be proof of intent or motive.138

In the 2008 Sprint/United Mgmt. Co. v. Mendelsohn case, the Supreme Court held that the admissibility of me-too evidence “depends on many factors, including how closely related the evidence is to the plaintiff’s circumstances and theory of the case”139 by virtue of factors such as similarities in the treatment of other employees and the plaintiff.140 Lower courts have treated Sprint as creating a narrow exception to the presumption that prior act evidence is inadmissible, and have sometimes excluded even me-too evidence that meets the Sprint criteria on other grounds, finding that the probative value of such evidence is outweighed by unfair prejudice141 or where it poses a danger of creating a “trial within a trial.” 142 One court stated that “more often than not, ‘me too’ evidence is not admitted at trial.”143

In response to me-too evidence, defendants have increasingly produced rebuttal witnesses to testify to the defendant’s respectful treatment of women. In theory, the same principles guide the admissibility of me-too and not-me-too evidence. In the non-sex harassment discrimination cases, this equivalence might make sense, since me-too and not-me-too evidence could plausibly deserve equal weight. If a plaintiff produces examples of other women paid less than they deserve, a defendant should surely be permitted to show that women on average are paid as well as men. Pay discrimination might not be universal, but is generally directed towards a group. In contrast, a sex harasser might plausibly treat women generally well but single out a small set of victims, such as those who seem easier prey by virtue of circumstance or temperament. To compound the problem, courts seem to apply the Sprint rules more leniently to not-me-too evidence, virtually always holding it admissible,144 while sometimes excluding me-too evidence.145

Judicial preference for not-me-too evidence poses a looming threat to sex harassment plaintiffs, and it demands a strong narrative in response. The strong misogyny theory cannot supply that narrative. A theory of sexual harassment that treats sexual behavior as a specific manifestation of generalized animus has tremendous difficulty explaining why a man who treats women generally well should single out a specific woman for misogynistic abuse in the form of sexual behavior. What is needed is a new narrative that can treat sex harassment as a gender-based wrong without characterizing it as a form of indiscriminate misogyny. The next Section outlines how such a new narrative might be constructed.

IV. New Norms of Sexual Harm

Propelled by dominance theory, the American law of sexual harassment took the momentous step of recognizing nonviolent sexual harm to adult women without invoking the norm of chastity. At the same time, the hybrid misogyny narrative behind those legal rules views harassment as the conduct of a small group of toxic misogynists, a deeply flawed picture that has produced deeply flawed doctrine.

Sexual misconduct is regarded by traditional sexual morality and first-wave feminism as an offense against chastity; by sexual autonomy theory as an offense against consent; and by dominance theory as an offense against women’s equality. Better legal doctrines require new narratives that build on the strengths of each of these models while learning from their limitations.

A. Consent and Its Limits

American society places tremendous value on individual autonomy and freedom of choice. A central premise of our moral thinking is the harm principle: only harm to third parties justifies interference with individual decisions and interactions between consenting adults. The traditional law of sex deviated from these core principles by scrutinizing consent far less in sexual settings than in other settings and by placing little value on women’s sexual autonomy. But in the past half century, sex law has become more consistent with these other broadly-held values.

Consent has thus become the touchstone of the American approach to the legal regulation of sex. By the later twentieth century, the resulting laws and norms had improved women’s sexual and economic freedom but left many women feeling unprotected from sexual predation. Many felt abandoned not only by the society against whose unequal institutions they struggled but also by liberal feminism, which sometimes viewed complaints about sexual harm as a step backwards from hard-won sexual freedom and towards a new neediness, suffused with a neo-Victorian rejection of female sexual pleasure.

Into this vacuum came dominance theory, which allowed women to protest sexual imposition without appearing querulous and wounded. Its rapid success in driving reform, however, had costs. It did to some extent revive stereotypes of female sexual coldness. It painted far too bleak a picture of male psychology and of a society that was rapidly improving its treatment of women. It is now a half century later, and a new theory of harm is needed, one that extends earlier autonomy models without the drawbacks of dominance theory.

The legal system’s notion of autonomy is primarily “thin.” Thin autonomy requires only that agents be free from wrongful interference with choice, without consideration of their actual capacity to act on this freedom.146 Autonomy theory becomes “thicker” as it builds in more requirements that consent be meaningful, and these extensions are an inherently value-laden exercise in defining new entitlements.147 Correspondingly, the concept of coercion (a violation of autonomy) expands as entitlements increase—coercion cannot be understood except against the background of entitlements.

The two major reform movements before dominance theory relied in part on autonomy arguments, wholly in the case of rape law reform and partly in the case of age of consent reform. In both instances, they required an autonomy theory that, if not maximally thin, was at least no thicker than that embodied in other areas of law. Age of consent reformers placed great weight on the comparison between sex law and the law of contracts and property. If minors could not enter into a valid sale of personal property, it seemed only reasonable to limit their capacity to consent to sexual relations. More recently, autonomy theorists have made a compelling case that sex law provides less protection than analogous law governing theft or professional conduct.148 For present purposes I will equate thin sexual autonomy with sexual autonomy based on entitlements found in non-sex areas of the legal system, although that is not entirely accurate, since those areas may already embody some thickness.

Thin sexual autonomy does not support extending the law of sexual harassment, because the American legal system simply does not provide enough legal protection to employees upon which to build. Employers are relatively free to mistreat employees, whose main redress is to leave, an option facilitated by the relative mobility of the American labor market. The tort of intentional infliction of emotional distress is available to employees only in the most extreme cases. The supervisor-employee relationship is not subject to even remotely the level of scrutiny applied to relationships such as therapists and patients or adults and minors. Any new narrative of sexual harm must capture how sex differs from property, contracts, and other areas in which the law supervises exchange relations.

To see how the model of thin autonomy plays out, consider the following hypothetical. A male supervisor, J, conducts the onboarding process for each new employee. After filling out paperwork, J provides an overview of company procedures. At the end, he tells many of the female new hires that he enjoys a quick hook-up after work, and that if she ever feels so inclined she should come to his office at the end of the day and they can repair to his place for the evening. He adds that this is completely voluntary and won’t affect her job. He does not ask for a response, never mentions the subject again unless the employee does, and neither rewards those who accept his offer nor sanctions those who do not.

Not everyone will characterize this scenario, by itself, as sexual harassment, though I would, but it surely at least provides evidence that could be part of a hostile environment case. But the thin autonomy model cannot easily even support its use as evidence. From its perspective, all that matters is that consent has been requested and any refusal honored. J has wholly met these requirements—his acts are not even a step in a troublesome direction. Professor Stephen Schulhofer, the author of a fascinating and important application of the autonomy paradigm to sexual harm,149 suggests that autonomy theory can be modified to acknowledge the problematic nature of J’s behavior.150 He proposes extending the unwelcomeness requirement of current hostile environment doctrine to allow for more consideration of circumstances, noting that that in “many of the reported cases, a supervisor confronted his female subordinate with a crude, impersonal sexual proposal. It seldom seems plausible to think that the woman was delighted by the idea or that only reticence prevented her from suggesting such an encounter herself.”151 In such circumstances, courts should presume unwelcomeness. In other words, unless the subordinate’s actions had somehow rebutted the presumption, the conduct would be evidence of harassment. In contrast, courteous advances of a personal or romantic character should not be presumed unwelcome, although a single gentle refusal should be sufficient to indicate that any future advances are unwelcome.152

The distinction between crude impersonal advances and romantic personal ones is onto something critically important that is not part of current doctrine and that goes a long way to describing why most people would consider J’s behavior disturbing. Precisely why this distinction is important demands further explanation. Autonomy theory typically honors the freedom to make offers. Schulhofer constructs an important new category in autonomy theory: lack of consent even to receive an offer.153 To this point his account is consistent with autonomy theory, which would typically honor an individual’s explicit refusal to entertain offers. But his next move is more complex: he suggests that the nature of a crude impersonal offer constitutes a proxy for lack of consent. Autonomy models generally disallow offers only if the counterparty is not legally competent to accept because, for example, she is a minor. Schulhofer does not advocate the complete legal incapacitation of subordinates, suggesting that a complete ban on relationships between supervisors and subordinates would intrude too much on the freedom of mature adults, who not infrequently choose to enter such relationships.154 And the special status that Schulhofer rightly accords to personal and romantic offers is not a pure proxy for welcomeness, as autonomy theory would require. An indication of interest from a physically attractive supervisor is more likely to be welcome than one from a less fortunate colleague, but neither Schulhofer nor anyone else would suggest that that should be legally relevant.

Not only the nature of the offer but the existence of a supervisory relationship clearly affects our view of J, but again thin autonomy theory cannot clearly explain why. Much of sex law reform has focused on instances in which victims of sexual harms received less protection than victims of harms to other interests such as property. But treating J’s proposal as presumptively unwelcome would extend the protection of sexual autonomy past the protection of property or other rights. Suppose J had told new employees that he was an Amway distributor and they should consider him for their wellness and cleaning product needs. This would be distasteful, and a corporate employer might well prohibit such behavior in the event that it became common. Yet no court would invalidate such a purchase by a subordinate on grounds of duress. By regarding J’s behavior as legally suspect, Schulhofer (and I and others) are proposing to extend the protection of sex law beyond that provided by the law governing other areas.155

Sex is different. Something about sex makes the supervisory setting more problematic for a request for sexual interaction than for a request for a financial transaction or for non-sexual social engagement. Something about sex gives special valence to the respectful or crude quality of the request, making even certain offers presumptively objectionable. But what is different?

B. Beyond Thin Autonomy

Autonomy theory is the dominant American approach to moral questions, but thin autonomy does not protect women from second-generation sexual harassment. What thicker model of autonomy, enriched by appeal to other values and entitlements, can do better? The first-wave feminist attack on the double standard ultimately rested too directly on Christian ideals of chastity to be straightforwardly imported into today’s secular legal system. Dominance theory’s powerful narrative was too dark to have broad public appeal, and from a theoretical point of view it contained no explanation of why men would seek to dominate women or why they might choose sex as their means to this end.

The harm principle upon which autonomy theory rests is ultimately empirical in implementation. Whether given actions cause harm, and of what kind, is a question on which evidence can be brought to bear, and I will now sketch out some possible approaches to developing an empirically grounded theory of sexual harm that might augment autonomy theory.

1. Sex is dangerous: a sex-neutral norm

Sex and the activities that surround it have consequences that extend far beyond erotic desire and satisfaction. To start the journey to thicker autonomy with this observation is to acknowledge that sexual harm is not always about misogynistic abuse, though it may be, and that fact is important. But sexual harm may also be about sex, or even about love, and is no less dangerous in these instances.

The sexual autonomy model seldom acknowledges the power of sexual emotions, and does so almost always outside the context of legal harms. Advocates of stronger norms against casual sex often note that sex can lead to strong feelings of emotional attachment. These feelings of attachment could be relevant in a number of ways to the law of sexual harassment,156 but I will focus on the three fact patterns identified earlier, which seem most likely to command a consensus in favor of extending the law. All three raise issues that touch on the many emotions besides desire that can arise in sexual settings.

When sexual advances are rejected or relationships are ended, the rejected party may experience pain, shame, anger, resentment, or feelings of inadequacy. A rejected supervisor may thus have difficulty treating a subordinate fairly. Fearing this, a subordinate who wishes to reject or end sexual contact may suffer great anxiety and or feel intimidated into sex even where there is no direct threat of harm. Rejected parties may feel this full range of emotions regardless of their initial intent: rejection is not fun whether advances were motivated by love, lust, or animus. Essentially the same emotions may be triggered when the problematic behavior is not a sexual advance but more indirect behavior of the kind described in Baskerville.

Equally serious though less obvious problems can arise when the supervisor is the rejecting party. The rejector may wish to avoid the emotions of the rejected; fear awkward situations; have a jealous spouse; or simply want a clean break. For any of these reasons, a supervisor who terminates a relationship may feel unable to return to a normal supervisory relationship, with detrimental effects to the subordinate’s career.

That love sometimes hurts is hardly a new insight—it’s hard to imagine popular music without it—but it has been curiously absent from discussions of sexual harassment. Perhaps the focus on misogynistic animus has been so single-minded that other types of danger have been ignored. Or perhaps the ethos of sexual freedom has made any allusion to the emotional dimension of sex seem vaguely old fashioned and puritanical. Over fifty years have passed since the advent of the sexual revolution, and it is now time to discuss these issues more honestly, with assistance from the growing body of empirical research on the emotional consequences of sex.

The strength of sexual emotions implies that truly free sexual choice requires more safeguards than truly free economic choice, especially in situations of unequal power such as the workplace. This observation, which I will call the dangerous-sex model, leads to a thicker autonomy model, which, unlike the misogyny model, supports protection even from harms not motivated by animus.

The dangerous-sex approach addresses some of the problem cases discussed earlier. Certainly it suggests that supervisory status be given much greater significance than it receives at present, in turn suggesting a different result in the jilted lover cases. Indeed, it identifies the problems in those cases as central to the understanding of the harm in sexual harassment: any of the formidable emotions surrounding sex, not just the malign ones, render the supervisory relationship dangerous. It refocuses the issue in the me-too evidence cases. The plaintiff is not trying to prove a general disposition to misogynistic behavior but to show a pattern of sexual behavior that is not necessarily manifested towards all women. The agency liability cases are more of a puzzle, since they make little sense even on their own terms. Of all three anomalies, they are most open to the complaint that they protect plaintiffs less from sexual harm than from other harms. In no other situation, including racial harassment, has Title VII’s policy of reconciliation been used to require internal reporting. Still, the misogyny narrative may have contributed to this wrong turn by creating an image of sex harassers as so anomalously malevolent, so outside the normal range of behavior, that their peers could be relied on to recognize and respond appropriately to their offenses. The dangerous-sex model instead emphasizes that essentially ordinary feelings and behavior become menacing when introduced into an environment of power. From this perspective it is unrealistic to expect self-policing by employers. Supervisors are, almost by definition, worth more to employers than those they supervise, since they are paid more, and employers have an incentive to favor supervisors over their subordinates in disputes between the two. This favoritism may be reinforced by the stronger social and collegial ties that exist between people at the same hierarchical level. The current agency liability rules seem to assume that harassers are so seriously pathological that their colleagues can overcome the strong forces that make people reluctant to find against valued colleagues and friends. But if sexual harm can result from normal behavior at the wrong time and place, the insistence on internal dispute resolution seems wholly misguided.

2.Developing new feminist norms

The dangerous-sex model is sex-neutral not only in terms of the legal rules it suggests, which do not differentiate between men and women, but in its assumptions about male and female sexuality. In American law and society, neutral legal rules are the preferred approach to promoting sex equality, and American feminists have attempted to avoid building assumptions about sex differences into their policy analysis.

At some point, however, it may be worthwhile to consider differences between men and women in the consequences of sexual activity, and even to entertain the possibility that these differences are not entirely environmental in origin. Neither the autonomy nor dangerous-sex models make it easy to claim sexual misconduct as inherently a feminist concern, since both take fundamentally sex-neutral perspectives on sexuality. Only the historically contingent fact of male economic power, itself unexplained, makes harassment of more importance to women than men. Both the autonomy and dangerous-sex models avoid the excessive pessimism of the dominance model but both have the opposite flaw. They do not explain why women are more likely to be disturbed than men by sexual harassment, and they answer the question of why sexual harassment is sex discrimination with only the wholly formal answer that the plaintiff would not have been treated as she was were it not for her sex.

A number of theories might provide an account of difference. I discuss only one that has recently not received the attention it deserves. This approach focuses on differences in male and female reproductive roles, and is not new to feminist theory. It can be found as early as Friedrich Engels157 and in several important second-wave feminists including Simone de Beauvoir,158 Gerda Lerner,159 and Shulamith Firestone.160 It is an approach that, though value-laden, is based on empirical observations about human behavior, as I believe future theories should be. This approach might be called the means of reproduction model, an allusion to the idea that patriarchies oppress women in order to give men control of the means of reproduction.

Women get pregnant and men do not. The shadow of forced pregnancy falls across all potentially coercive heterosexual interaction even in a post-contraceptive era. Such fears explain in part why rape is traumatic in a way that other assaults are not. Perhaps it is less obvious why such intense fears can be triggered by situations in which the possibly of coerced intercourse is not imminent. Consider the case of Valerie Baskerville, discussed earlier, whose supervisor told her, inter alia, that an announcement over the work PA system meant “all pretty girls run around naked,” and who indicated that his wife told him that his sexual interest in Baskerville was becoming too serious.161 The threat of sexual coercion was in the air, but surely it was not an immediate possibility. Why was her situation more deserving of legal protection that that of an employee who suffers non-sexual abuse?

To say that the threat of forced pregnancy drives women’s sexual fear does not require that that fear result from careful calculation of the likelihood of pregnancy. Evolutionary psychologists suggest that we have two distinct mental tracks.162 Domain-general mechanisms give humans some capacity to respond to novel situations. These work alongside domain-specific mechanisms that have evolved by natural selection to respond to recurring adaptive problems of the environment inhabited by early humans. These domain-specific mechanisms operate not simply by telling us what to do, but by filling us with powerful emotions.163 Though we are capable of fearing things for which no domain-specific mechanism exists, our most primal reactions are ancient and domain-specific. Children instinctively fear snakes but not cars, although in modern life cars pose an incomparably greater risk to their safety. Domain-specific mechanisms are not finely calibrated to actual risk, and once triggered they are strong. Snakes in a glass cage or even in a photograph can inspire visceral fear, while cars do not. Similarly, the threat of sexual coercion is frightening even when victims know that the chances of actual coerced sex are small and the chances of resulting pregnancy are smaller.

Feminists become understandably nervous at this point. The suggestion that evolved dispositions play a role in sexual behavior conjures up the views of an early school of evolutionary psychology that might be called traditionalist, because it scientizes the traditionalist view of sex roles and it takes a restrictive view of female sexuality, grounded in part on empirical claims about women’s lower interest in sex.164 Later researchers such as Sarah Blaffer Hrdy,165 Jane Lancaster,166 and Barbara Smuts167 have presented a very different picture of human sexual predispositions that allows for far more variation between societies and a much more expansive and complex view of female sexuality. This sophisticated theory is consistent with, and even helpful to, a feminist perspective on sexual harassment.

All evolutionary psychologists agree on one foundational principle of sexual behavior, the theory of parental investment. Through pregnancy and its corollaries such a breastfeeding, human females have a much higher level of obligatory investment in each offspring. Women should be more discriminating in selecting sex partners, since the possibility of pregnancy make each copulation a greater potential commitment of resources for a female than for a male.168 Early evolutionary psychologists took this to mean that women were interested only in long-term relationships in which they traded sexual fidelity for male provisioning.169 Later researchers pointed out the errors in this last leap. The possibility of pregnancy means only that females should be more sexually selective than males and prefer partners of high genetic quality. It does not imply a female taste for monogamy or long-term relationships. Few mammal species are monogamous or have significant paternal provisioning, yet in all females are more sexually selective. High status males tend to pursue numerous partners, while high status females tend to seek better partners, though sometimes quite a few of them.170 Sharon Stone once remarked that one advantage of being famous was that “I find I get to torture a higher class of men.”171

The selectivity principle helps explain why coercive sex is more frightening to women than to men. Any act of coerced sex is potentially an act of coerced reproduction that could create an indestructible link between the victim and the coercer. Because sexual choice is of overwhelming importance to females, powerful fear can be triggered by non-copulatory sexual coercion, unexecuted threats of sexual coercion, intercourse without risk of pregnancy because of age or contraception, and situations in which the threat of coercion is not immediate.172 Just as in any other aspect of human behavior and physiology, there is a wide range of individual difference, but for the average woman, the possibility of sex is more fraught than for the average man not because of lack of sexual desire but because the potential consequences of sex are far more significant. The reproductive component of sex can provide an explanation for why sex is different and why we might protect sexual autonomy more than other autonomy interests. At the deepest emotional level, unwanted sex can never be just sex or just violence but is an act of reproductive coercion that simply has no analogue in any non-sexual behavior. Pressure to buy Amway products from a supervisor is uncomfortable, but pressure to have sex is terrifying.

Traditionalist evolutionary psychologists were wrong to leap from parental investment theory to the view that women seek only monogamous long-term relationships, but their account of male psychology provides a new perspective that can bridge traditionalist and feminist accounts of male domination. All evolutionary psychologists note that the long dependency of human infants means that paternal provisioning increases the likelihood of a child’s survival; that men are reluctant to support their children unless they can be certain about paternity; and that men highly value chastity in long term mates.173 Traditionalists and others mistakenly assumed that extensive male provisioning was a universal—in fact, the relative economic contribution of men and women varies greatly between societies. They were correct, however, that the sexual division of labor leaves most reproductive tasks to women; that on average men make a higher contribution to subsistence; and that societies in which men make a large contribution to subsistence are often organized to restrain female sexuality to ensure paternity certainty through a sexual double standard.174 Societies with a double standard vary in their requirements of female virtue, running the range from strict chastity requirements, often brutally enforced, to a looser expectation that women require men to display respect and seriousness before becoming sexually intimate. But even in relatively permissive societies, women are strongly stigmatized for crossing the elusive line between desirable hotness and repellent sluttiness, and the pain of this stigma is keenly felt.175

The consequences of being labelled a slut are serious. Women categorized as sluts occupy an exceedingly low rung on the social scale. Many men regard them as outside the class of women eligible for long-term serious relationships, and they are at much greater risk of sexual imposition—recall that until recently evidence of a women’s prior sexual experience was considered compelling evidence against a rape charge. Often the suggestion of sexual experience in a woman is understood to imply other negative personal traits. Thus, a supervisor who engages in a “crude, impersonal” sexual conduct towards an employee is indicating that she is a low status person who can be taken advantage of sexually and in other ways as well.176

The means-of-reproduction framework therefore provides a view of sexual harassment not just as a wrong but as a discriminatory wrong. It fills a key gap in the autonomy model by providing a reason for extending protection to sexual harms past the level provided to property wrongs. Like dominance feminism and first-wave feminism, means-of-reproduction feminism stresses role of sex in the social and economic control of women. Unlike these alternatives, it does so without resorting to norms of female chastity or assumptions about female sexual coldness. Women are more sexually vulnerable not because of their fragility or lack of desire but because the possibility of pregnancy makes coercive sex frightening and makes the suggestion of sexual promiscuity degrading in ways it would not be to a man. The means-of-reproduction approach provides an account of why men might wish to subordinate women that is both more empirically satisfying than prior theories and less gloomy, acknowledging the variation among societies and individuals and the possibility of movement towards more just social arrangements.

In the context of sexual harms, the means-of-reproduction model enlarges the perspective of the dangerous-sex model to clarify the special emotional consequences of sex for women. Women are more sexually vulnerable not because they are prudish or coy, but because sex has potentially far more significant consequences for them than it does for men. Society may help reduce that differential by expanding women’s reproductive rights, but human emotional responses are to some extent those of our ancestors in the environment of evolutionary adaptation. American law and norms favor formal equality, and the mean of reproduction perspective probably does not change the policy prescriptions of the dangerous-sex model. From a purely formal point of view, sex harassment is probably best thought of as a sex-neutral offense, a wrong because it would not have occurred but for the plaintiff’s sex. Sex differences are relevant because they inform the application of the Harris factors: whether a reasonable person in the plaintiff’s position would have found the behavior sufficiently severe, threatening, or humiliating to constitute a hostile environment.177

But narratives matter. Sexual neutral models perform important functions, but a complete picture requires something more. In practice, adopting a neutral perspective means a male perspective will determine the governing legal structures, and that has historically failed to protect women. Judge Posner could not imagine that he would have been upset by Hall’s behavior, and no doubt few men would have been. Narratives are needed not so much to provide doctrinal arguments as to enhance our ability to understand experiences beyond our own.

V. Conclusion

Subordination theory transformed the public and judicial view of sexual misconduct. Because its dark picture of pervasive misogyny was never broadly accepted, it metamorphosed into a meme of a small class of misogynists, driven by the need to dominate and control women, and deserving of both ostracism and legal sanctions.

This powerful image of harassers enabled the law to expand liability for sexual harassment past the narrow confines of explicit quid pro quo, but at the price of distorting the doctrinal elaboration that followed. Sexual harassment came to be seen as essentially a form of rape. Sometimes, as in Meritor, it was. But serious sexual harm can also result from motives and behavior much closer to normal sexual conduct, a possibility obscured by the conflation of harassment and rape. If harassers were rapists, the coercive nature of the supervisory relationship needed no special recognition; most organizations could be counted on to self-police; defendants should be able to defend their character; and retaliation motivated by hurt feelings should be not actionable. All sexual harm was rape, committed by rapists.

Of course, not all sexual harm is the equivalent of rape, and not all men who cause sexual harm are the moral equivalent of rapists. The challenge now facing public policy is the regulation, whether by law or norms, of a vast gray area of motive and behavior. What kinds of harm short of that suffered in a violent assault should the law remedy? What is the relevance of motive? Some sexual harms will be unavoidably outside the law, but some intermediate harms are deserving of legal relief, especially those that occur in environments of unequal power. A female employee may be seriously injured by the behavior of a supervisor who is immature or woundedbut not malicious. As long as the misogynynarrative prevailsnuanved discussion of intermediate sexual harm is impossible. New narratives, such as those that stress the emotional dangerousness of sex, and perhaps even its special risks for women, are needed.

New norms, however, require a shift in the rhetoric of culpability. If women are to be protected from sexual harm that is significantly short of rape, they cannot claim that that harm is equivalent to that of rape, or that it is inflicted by men who are essentially rapists. In a pathbreaking article twenty-five years ago, Professor Linda Krieger argued that discrimination, in the most general sense, was frequently motivated not by animus but by a variety of unconscious biases. The resulting discrimination, she noted, was “unintended and, for many people, earnestly undesired.”178 While arguing that the law should provide remedies for unconscious discrimination, she cautioned against applying the same level of moral condemnation to those who committed unconscious discrimination as to those who engaged in the paradigmatic deliberate variety. Inappropriate levels of censure, she predicted, would backfire, heightening tension and creating resistance.179

New narratives of sexual harm require similar modulation in the rhetoric of blame. The dangers of overstating culpability can be seen all too clearly in the #MeToo moment. Astonishingly to many (or at least to many men), the #MeToo moment showed the ineffectiveness of the current legal system. At the same time, the dangers of a dichotomized view of male behavior were far more evident in the Twittersphere than they had been in the courtroom. In court, narratives of sexual harm are a subtle influence on the logic of the law, always present but not easily detected. In social media and on the Internet more generally, moral narratives are always front and center. The initial wave of allegations against individuals such as Harvey Weinstein involved behavior that was either rape or an extreme abuse of power whose immorality few questioned. The narrative of misogyny was rightly used in this setting. As the #MeToo movement progressed, new charges continued to raise issues central to the protection of women’s equality, but the conduct described became less extreme and fit less readily into the models of rape and misogyny. The behavior in question often involved suggestive language or touching. Some of the accused, like Garrison Keillor, may have been impelled, as Judge Posner said of Michael Hall, by immaturity rather than by animus. The discussion shifted from professional settings, where the element of power transformed the creepy into the coercive, to the purely social, culminating in the claims made against Aziz Ansari involving callous but lawful conduct during a date. Yet these men received the same heavy artillery accusations of misogyny as did Weinstein.180

Some commentators, including some feminists, noted that #MeToo disregarded crucial distinctions between widely differing behavior. Some critics went further, suggesting that concerns about anything short of the most egregious conduct would rob women of their agency and even “strip sex of eros.”181 The view that the line must be drawn narrowly, to encompass only the most egregious behavior, was explicitly tied to the misogyny theory: “Shouldn’t sexual harassment imply a degree of hostility?”182

#MeToo supporters responded that no-one was actually equating leering sexual advances with rape. In some sense this response was correct. If pressed, most #MeToo advocates would no doubt agree that not all charges were equally serious. Yet there has been no sustained discussion of the many gradations of undesirable sexual conduct. And as Krieger predicted, condemnation in excess of what is warranted has contributed to backlash. Some contend that #MeToo is an effort to create division between men and women,183 while others insist that if they must worry about being accused they will simply avoid women professionally.184 Not all of these reactions deserve sympathy, but some seem to me the result of genuine confusion and resentment of a world in which the rules seem unclear and the penalties for transgression arbitrary.

I have done little here to provide a practical guide for the perplexed on the specific categories of intermediate sexual harm. My more modest goals have been to make the case for moving away from the strong misogyny model and to suggest some paths that journey might take. Without new and less morally charged narratives of harm, there can be no discussion of how the law and social norms can make modulated assessments of the culpability of those who cause sexual harm, and provide protection against significant sexual harms that are not motivated by misogyny.

  • 1Stephen J. Schulhofer, Unwanted Sex: The Culture of Intimidation and the Failure of Law 17–47 (1988). See also Jed Rubenfeld, The Riddle of Rape-by-Deception and the Myth of Sexual Autonomy, 122 Yale L. J. 1372, 1391–92 (2013).
  • 2Josephine Butler, Social Purity, in The Sexuality Debates 170 (Sheila Jeffreys ed., 1987).
  • 3Martin S. Weinberg, Rochelle Ganz Swensson & Sue Kiefer Hammersmith, Sexual Autonomy and the Status of Women: Models of Female Sexuality in U.S. Sex Manuals from 1950 to 1980, 30 Soc. Problems. 312, 312–13 (1983).
  • 4In this Article, I focus on the limits of the misogyny paradigm in sexual harassment law. Katharine Baker has similarly argued that even rape law has been distorted by overstating the malignancy of rapists’ motives. Katharine K. Baker, Once a Rapist? Motivational Evidence and Relevancy in Rape Law, 110 Harvard Law Review 563 (1997).
  • 5Meritor Sav. Bank, FSB v. Vinson, 477 U.S. 57, 66 (1986).
  • 6Oncale v. Sundowner Offshore Serv., Inc., 523 U.S. 75, 80 (1998).
  • 7See, e.g., Kimberly Kessler Ferzan, Consent and Coercion, 50 Ariz. St. L.J. 951, 951–92 (2018) (proposing a “normative impairment” definition of coercion that focuses on both the “blameworthiness of the coercer” and the “involuntariness of the consenter’s choice”); Rubenfeld, supra note 1, at 1412 (pointing out the incomplete definition of coercion and suggesting supplementing it with consideration of whether sexual activity involved “deception”); Michal Buchhandler-Raphael, Criminalizing Coerced Submission in the Workplace and in the Academy, 19 Colum. J. Gender & L. 409, 437–38, 442 (2010) (advocating for a model of sexual coercion that considers a variety of factors, i.e., economic inequality).
  • 8Title VII of the Civil Rights Act of 1964, 42 U.S.C.S. §2000e–2(a) (1964).
  • 9I speak in this article primarily of sexual harms to women. This focus is not to minimize the importance of protections against sexual harms to others, including men and gender nonconformists. However, the relation of these to harms to women is complex and beyond the scope of this article.
  • 10Gwen J. Broude, Extramarital Sex Norms in Cross-Cultural Perspective, 3 Behavior Science Research 181, 182 (1980).
  • 11See generally Keith Thomas, The Double Standard, 20 J. of the Hist. of Ideas 195 (Apr. 1959) (examining the evolution of the double standard).
  • 12Id.
  • 13At a minimum, female chastity was needed to ensure certainty of paternity, a condition of inducing men to care for their children. David Hume, A Treatise of on Human Nature 331–32 (T.H. Green & T.H. Grose eds., Longmans, Green & Co. 1874) (1738). Hume, who is sometimes regarded as a protofeminist, seems to have regarded the double standard as necessary but unfair.
  • 14The choice between the terms “prostitute” and “sex worker” is a difficult one. The former carries strong and undesirable connotations of disgrace and moral blame, and to avoid these some Victorian reformers substituted the term Magdalenism. However, the generally preferable “sex worker” is anachronistic in contexts that describe the attitudes of earlier periods. My choice of terminology reflects this tradeoff.
  • 15Adultery, for example, was typically grounds for divorce for the wronged husband but not the wronged wife, and in some circles the failure of a married man to keep a mistress was regarded as unmanly. Thomas, supra note 11, at 195, 199.
  • 16Id. at 198.
  • 17Mary E. Odem, Delinquent Daughters: Protecting and Policing Adolescent Female Sexuality in the United States, 1885–1920 (University of North Carolina Press, 2d ed. 1995).
  • 18Schulhofer, supra note 1, 1917 (1988).
  • 19Id. at 18.
  • 20Id. at 18–20.
  • 21Id. at at 18–19.
  • 22Vivian Berger, Man’s Trial, Woman’s Tribulation: Rape Cases in the Courtroom, 77 Colum. L. Rev. 1, 13–18 (Jan. 1977).
  • 23Coventry Patmore, The Angel in the House (1854).
  • 24John Ruskin, Of Queens’ Gardens, in Sesame and Lilies (H. Bloom ed. 1983) (1910).
  • 25Jane Addams, Women and Public Housekeeping (1910). In this view, virtually all of domestic policy, in the sense of non-foreign policy, was in fact domestic, in the sense of women’s work.
  • 26Butler, supra note 2, at 172-74. Jane Larson observed that these reformers “saw sexuality as a vehicle of power that in complex ways kept women subordinated in society. In response, they created a vigorous sexual politics that challenged not just private, but also public power. Ultimately, they questioned the state’s conferral of privilege in law of male sexual interests to the detriment of women and girls; they thus exposed the state’s complicity in what otherwise appeared to be wholly private acts of sexual oppression.” Jane E. Larson, “Even a Worm Will Turn at Last”: Rape Reform in Late Nineteenth-Century America, 9 Yale L.J. & Human. 1, 4 (1997).
  • 27Id. at 27; Reva B. Siegel, Introduction: A Short History of Sexual Harassment, in Directions in Sexual Harassment Law 12–13 (Catherine A. MacKinnon & Reva B. Siegel eds., 2004); Emma Goldman, The Traffic in Women, in Anarchism and Other Essays 177 (1910).
  • 28Suffrage is often seen as the first step in the emancipation of women when it was in fact as much the culmination of a broad variety of reform efforts undertaken by a women’s movement comprised of coalition of diverse views. Jane E. Larson, “Even a Worm Will Turn at Last”: Rape Reform in Late Nineteenth-Century America, 9 Yale J.L. & the Human. 1, 3 (1997) (“[F]eminism, evangelism, maternalism, domesticity, and moral reform energized the mainstream of the woman movement.”). See also Sheila Jeffreys, The Spinster and Her Enemies: Feminism and Sexuality 1880–1930 chs. 3–4 (London: Pandora, 1985).
  • 29Odem, supra note 17, at 37.
  • 30Larson, supra note 26, at 1, 8–10.
  • 31Brian Donovan, Gender Inequality and Criminal Seduction: Prosecuting Sexual Coercion in the Early-20th Century, 30 Law and Soc. Inquiry 61, 66 (2005). See generally H. W. Humble, Seduction as a Crime, 21 Colum. L. Rev. 144 (1921); Jane E. Larson, “Women Understand so Little, They Call My Good Nature ‘Deceit’”: A Feminist Rethinking of Seduction, 93 Colum. L. Rev. 374 (1993); Lea VanderVelde, The Legal Ways of Seduction, 48 Stan. L. Rev. 817 (1996); Sara McLean, Confided to His Care or Protection: The Late Nineteenth-Century Crime of Workplace Sexual Harassment, 9 Colum. J. Gender & L. 47 (1999).
  • 32Mo. Rev. Stat. §1260 (1879).
  • 33Siegel, supra note 27, at 11–12.
  • 34D. R. N. Blackburn, Criminal Seduction, 31 Cent. L.J. 44, 44–50 (1890).
  • 35See, e.g., Margaret Sanger, Women and the New Race 226, 229 (1920).
  • 36Emma Goldman, The Traffic in Women, in Anarchism and other essays (1910).
  • 37Latex was invented in 1916, greatly improving both the reliability and the experience of using barrier methods such a condoms and diaphragms. See Hallie Lieberman, A Short History of the Condom, JSTOR Daily (June 8, 2017), [].
  • 38Alfred Kinsey, Sexual Behavior in the Human Male (1948); Alfred Kinsey, Sexual Behavior in the Human Female (1953).
  • 39Wilhelm Reich, The Sexual Revolution (1936).
  • 40See, e.g., Theodoor Hendrik van de Velde & Margaret Smyth, Ideal Marriage, Its Physiology and Technique (1928).
  • 41See The Playboy Philosophy, Part I, Installments 1­–7 (1962), [].
  • 42See Helen Gurley Brown, Sex and the Single Girl (1962).
  • 43Stephen J. Schulhofer, Taking Sexual Autonomy Seriously: Rape Law and Beyond, 11 Law & Phil. 35, 92–93 (1992).
  • 44Rubenfeld, supra note 1, at 1342.
  • 45See, e.g., Wash. Rev. Code §18.130.180(24) (1984) (prohibiting any licensed health professionals from engaging in “sexual contact with a client or patient”); Cal. Bus. & Prof. Code §6106.9 (1993) (prohibiting sexual relationships between lawyers and clients under various circumstances); see also Cotton v. Kambly, 300 N.W.2d 627, 629 (Mich. App. 1981) (sexual relationship between psychiatrist and patient constitutes malpractice); McDaniel v. Gile, 230 Cal. App. 3d 363, 376 (1991) (describing a lawyer’s threat to withhold legal services if client did not comply with sexual demand constitutes malpractice).
  • 46Tomkins v. Pub. Serv. Elec. & Gas Co., 422 F. Supp. 553, 557 (D.N.J. 1976), rev’d, 568 F.2d 1044 (3d Cir. 1977) (“If the plaintiff’s view were to prevail, no superior could, prudently, attempt to open a social dialogue with any subordinate of either sex. An invitation to dinner could become an invitation to a federal lawsuit if a once harmonious relationship turned sour at some later time. And if an inebriated approach by a supervisor to a subordinate at the office Christmas party could form the basis of a federal lawsuit for sex discrimination if a promotion or a raise is later denied to the subordinate, we would need 4,000 federal trial judges instead of some 400.”); Miller v. Bank of America, 418 F. Supp. 233, 236 (N.D. Cal. 1976) (“It is conceivable, under plaintiff’s theory, that flirtations of the smallest order would give rise to liability. The attraction of males to females and females to males is a natural sex phenomenon and it is probable that this attraction plays at least a subtle part in most personnel decisions. Such being the case, it would seem wise for the Courts to refrain from delving into these matters short of specific factual allegations describing an employer policy which in its application imposes or permits a consistent, as distinguished from isolated, sex-based discrimination on a definable employee group”); Corne v. Bausch & Lomb, Inc., 390 F. Supp. 161, 163(D. Ariz. 1975) (“[A]n outgrowth of holding such activity to be actionable under Title VII would be a potential federal lawsuit every time any employee made amorous or sexually oriented advances toward another.”).
  • 47See Barnes v. Costle, 561 F.2d 983, 996 (D.C. Cir. 1977) (explaining that the Restatement imposes a narrow test for vicarious liability—“The tort must be one accomplished by an instrumentality, or through conduct associated with the agency status”); Williams v. Saxbe, 413 F. Supp. 654, 656–61(D.D.C. 1976).
  • 48Susan Brownmiller, Against Our Will: Men, Women and Rape 377–78 (1975).
  • 49“Sexual harassment, most broadly defined, refers to the unwanted imposition of sexual requirements in the context of a relationship of unequal power. Central to the concept is the use of power derived from one sphere to lever benefits or impose deprivations in another.” Catherine MacKinnon, Sexual Harassment of Working Women: A Case of Sex Discrimination 1 (1979).
  • 50Brownmiller, supra note 46.
  • 51Catharine A. MacKinnon, Toward a Feminist Theory of the State 174, 175 (1989).
  • 52MacKinnon, supra note 45, at 1; Andrea Dworkin, Intercourse (1987).
  • 53Louise Thistle, Sexual Harassment in the Workplace: A Struggle for Dominance, San Diego Union, Apr. 20, 1980, at 58; Karen Kirk, Fighting Abuse at Work, Columbus Dispatch, Oct. 10, 1980, at 46.
  • 54Patricia Holt, MacKinnon’s Feminist View: Embattled and Unmodified, San Francisco Chronicle, July 2, 1987, at 70.
  • 55Id.
  • 56See, e.g., Jerry Barnett, Is Anti-Sex Feminism a Step Backwards for Women’s Rights?, FWSA Blog (Dec. 6, 2013), [].
  • 57561 F.2d 983 (D.C. Cir. 1977).
  • 58Id. at 990. The critical role of but-for causation has since been reaffirmed by the Supreme Court in Oncale, 118 S. Ct. at 1002.
  • 59MacKinnon, Sexual Harassment of Working Women, supra note 45, at Ch. 3–4.
  • 60Brief of Respondent, at 30, Meritor Savings Bank, FSB v. Vinson, 477 U.S. 57 (1986), 1986 U.S. S. Ct. Briefs LEXIS 102.
  • 61MacKinnon, supra note 45, at 40–41.
  • 62Intent to Conduct Public Scoping Meeting in Compliance with the National Environmental Policy Act, 45 Fed. Reg. 74, 676 (1980).
  • 63“Harassment on the basis of sex is a violation of section 703 of title VII. Unwelcome sexual advances, requests for sexual favors, and other verbal or physical conduct of a sexual nature constitute sexual harassment when (1) submission to such conduct is made either explicitly or implicitly a term or condition of an individual’s employment, (2) submission to or rejection of such conduct by an individual is used as the basis for employment decisions affecting such individual.” 29 C.F.R. §1604.11 (1980).
  • 64Id.
  • 65477 U.S. 57 (1986).
  • 66Brief of Respondent, supra note 56, at 1.
  • 67In particular, Vinson claimed that shortly after she was hired, Taylor:

    invited her out to dinner and, during the course of the meal, suggested that they go to a motel to have sexual relations. At first she refused, but out of what she described as fear of losing her job she eventually agreed. According to respondent, Taylor thereafter made repeated demands upon her for sexual favors, usually at the branch, both during and after business hours; she estimated that over the next several years she had intercourse with him some 40 or 50 times. In addition, respondent testified that Taylor fondled her in front of other employees, followed her into the women’s restroom when she went there alone, exposed himself to her, and even forcibly raped her on several occasions. These activities ceased after 1977, respondent stated, when she started going with a steady boyfriend.

    Vinson, 477 U.S. at 60.

  • 68Id. at 67 (quoting Henson v. Dundee, 682 F.2d 897, 902 (11th Cir. 1982)).
  • 69For criticism of this approach see Lynn T. Dickinson, Quid Pro Quo Sexual Harassment: A New Standard, 2 Wm. & Mary J. Women & L. 107 (1995); Marlissa Vinciguerra, Note, The Aftermath of Meritor: A Search for Standards in the Law of Sexual Harassment, 98 Yale L.J. 1717, 1718–19 (1989).
  • 70Harris v. Forklift Sys., 510 U.S. 17, 21–22 (1993).
  • 711) the frequency of the discriminatory conduct, 2) its severity, 3) whether it is physically threatening or humiliating or a mere offensive utterance; and 4) whether it unreasonably interferes with an employee’s work performance. Harris, 510 U.S. at 23.
  • 72“The abuse of authority by supervisors of either sex for personal purposes is an unhappy and recurrent feature of our social experience [P]laintiff’s theory rests on the proposition, with which this Court concurs, that the power inherent in a position of authority is necessarily coercive. Any subordinate knows that the boss is the boss whether a file folder or a dinner is at issue. If the plaintiff’s view were to prevail. An invitation to dinner could become an invitation to a federal lawsuit if a once harmonious relationship turned sour at some later time.” Tomkins, 422 F. Supp. at 557.
  • 73Harassment does not include “ordinary socializing in the workplace—such as male-on-male horseplay or intersexual flirtation.” Oncale, 118 S. Ct. at 1003.
  • 74“This standard takes a middle path between making actionable any conduct that is merely offensive and requiring the conduct to cause a tangible psychological injury. As we pointed out in Meritor, “mere utterance of an epithet which engenders offensive feelings in an employee,” ibiddoes not sufficiently affect the conditions of employment to implicate Title VII. Conduct that is not severe or pervasive enough to create an objectively hostile or abusive work environment—an environment that a reasonable person would find hostile or abusive—is beyond Title VII’s purview.” Harris, 510 U.S. at 21.
  • 7550 F.3d 428 (7th Cir. 1995).
  • 76Id. at 430.
  • 77Id.
  • 78Id.
  • 79Id.
  • 80Id. at 431.
  • 81Id.
  • 82Id.
  • 83Id.
  • 84Id.
  • 85Rebecca Hanner White & Linda Hamilton Krieger, Whose Motive Matters?: Discrimination in Multi-Actor Employment Decision Making, 61 La. L. Rev. 495, 501 (2001) (“For years, it has (or should have) been clear that discriminatory intent or motive is not coextensive with hostile animus”).
  • 86My personal favorite is the Sexual Harassment episode. The Office US: Sexual Harassment (NBC television broadcast Sept. 27, 2005), part of which can be seen at, Michael Scott Fights Corporate – The Office US, YouTube (Jan. 30, 2015), [].
  • 87See generally Baskerville, 50 F.3d 428.
  • 88524 U.S. 742 (1998).
  • 89524 U.S. 775 (1998).
  • 90Burlington, 524 U.S. at 751–53.
  • 91Restatement (Second) of Agency §219(1).
  • 92The Restatement is not entirely consistent in its definition of scope of employment, at one point appearing to require that action be motivated in part by a desire to serve the master (§228) and at another simply that it be authorized or incidental to authorized conduct (§229). Id. §§228–29. Some employers seemed to regard sexual access to subordinates as a perquisite of status. Julie Berebitsky, Sex and the Office: A History of Gender, Power, and Desire 144 (2012). At such employers, harassment might be said to be authorized.
  • 93Burlington, 524 U.S. at 758–59.
  • 94Id. at 760.
  • 95Id. at 763.
  • 96Id.
  • 97Id. at 763–64.
  • 98“For example, Title VII is designed to encourage the creation of antiharassment [sic] policies and effective grievance mechanisms. Were employer liability to depend in part on an employer’s effort to create such procedures, it would effect [sic] Congress’ intention to promote conciliation rather than litigation in the Title VII context.” Id. at 764.
  • 99Id. at 765.
  • 100Id.
  • 101Louise F. Fitzgerald, Suzanne Swan & Karla Fischer, Why Didn’t She Just Report Him? The Psychological and Legal Implications of Women’s Responses to Sexual Harassment, 51 J. Soc. Issues 117, 127 (1995) (60% of non-reporters believed they would be blamed for the incident if they made a formal complaint; 60% believed complaints would be ineffective because nothing would be done); Chelsea R. Willness, Piers Steel & Kibeom Lee, A Meta-Analysis of the Antecedents and Consequences of Workplace Sexual Harassment, 60 Personnel Psychol. 127 (2007); David Sherwyn, Michael Heise & Zev J. Eigen, Don’t Train Your Employees and Cancel Your “1-800” Harassment Hotline: An Empirical Examination and Correction of the Flaws in the Affirmative Defense to Sexual Harassment Charges, 69 Fordham L. Rev. 1265, 1290–92 (2001).
  • 102Lilia M. Cortina & Vicki J. Magley, Raising Voice, Risking Retaliation: Events Following Interpersonal Mistreatment in the Workplace, 8:4 J. Occupational Health Psychol. 247, 255 (2003).
  • 103See Jackson v. Arkansas Dep’t of Educ., Vocational & Technical Div., 272 F.3d 1020, 1026 (8th Cir. 2001) (finding nine-month delay in reporting sexual harassment to be unreasonable); Shaba v. IntraAction Corp., No. 02 C 5173, 2004 U.S. Dist. LEXIS 78, at *16 (N.D. Ill. Jan. 5, 2004) (finding two-month delay in reporting sexually harassing conduct of supervisor—during which time employee recorded events in a log and talked to coworkers about the harassment—to be unreasonable); Dedner v. Oklahoma, 42 F. Supp. 2d 1254, 1260 (E.D. Okla. 1999) (finding three-month delay in reporting sexual harassment by supervisor, who had previously been fired for sexually harassing behavior and then reinstated, to be unreasonable); Phillips v. Taco Bell Corp., 83 F. Supp. 2d 1029 (E.D. Mo. 2000) (holding that plaintiff’s delay in reporting for three months after the first incident made her behavior unreasonable, even though she reported once behavior escalated); Kohler v. Inter-Tel Tech., No. C98-0378, 1999 U.S. Dist. LEXIS 5425, at *15–16 (N.D. Cal. Apr. 13, 1999) (finding that employees who do not report are almost always found to have acted unreasonably).
  • 104Clark County Sch. Dist. v. Breeden, 532 U.S. 268, 270–71 (2001) (per curiam).
  • 105Conatzer v. Med. Prof’l. Bldg. Serv., Inc., 255 F. Supp. 2d 1259, 1264–71 (N.D. Okla. 2003) (Plaintiff’s supervisor rubbed against the side of plaintiff’s chest on Sept. 28 and then placed her head in a headlock between his knees on October 11 or 12. On October 15, the plaintiff made a formal complaint under the employer’s sexual harassment policy. Even though the first incident occurred in front of another supervisor, the district court held that the employer’s failure to take any action until after the plaintiff made a formal complaint to be reasonable, because that single incident did not give the employer notice of the existence of a hostile environment requiring correction. However, the district court concluded that plaintiff’s failure to make a formal report immediately after that incident, despite a formal incident 3–4 days after the second incident, constituted an unreasonable failure to take advantage of preventative or corrective opportunities provided by the employer).
  • 106Anderson v. Leigh, No. 98 C 50169, 2000 U.S. Dist. LEXIS 1584, at *17–19 (N.D. Ill. Feb. 10, 2000).
  • 107Id. at *3–4, *22.
  • 108Childress v. PetsMart, Inc., 104 F. Supp. 2d 705, 707–09 (W.D. Tex. 2000).
  • 109Sconce v. Tandy Corp., 9 F. Supp. 2d 773, 775–76 (W.D. Ky 1998).
  • 110See Sherwyn, supra note 99.
  • 111Good data on sex and love at work is surprisingly hard to find. Most surveys that directly examine the topic appear to be done by commercial vendors of human resource related services such as Vault and CareerBuilder. The sample sizes are typically under 1000, which is problematic for phenomena that might well vary widely among sectors and regions. Academic studies on larger samples that consider adjacent topics, such as marital happiness, sometimes shed light on office relationships. Taken together, this body of research, however incomplete, does seem to provide relatively consistent results, at least as to the order of magnitude of various phenomenon. One survey found that 31 percent of workers who started dating at work eventually married. This survey was conducted online by The Harris Poll from November 28 and December 20, 2017 and included a representative sample of 809 full-time workers across industries and company sizes in the U.S. private sector. Rachel Nauen, Office Romance Hits 10-Year Low, According to Career Builder’s Annual Valentine’s Day Survey, CareerBuilder (Feb. 1, 2018), []. A representative sample of 19,131 individuals married between 2005 and 2012 found that 65.05% of relationships began offline and of those 21.66% began at work, implying that about 14% began at work. John T. Cacioppo et al., Marital Satisfaction and Break-ups Differ across On-line and Off-line Meeting Venues, 110 Procs. Nat’l Acad. Sci. U.S. (PNAS) 10135 (June 18, 2013). The marital satisfaction level of work relationships was somewhat lower at a significant level, though to my eye the effect size does not seem particularly large. Id. at 101373.
  • 112The 2018 Vault Office Romance Survey Results, Vault Careers (Feb. 12, 2018), [].
  • 113CareerBuilder surveys find over the years that between 36 and 41% of workers have ever dated a co-worker. Nauen, supra note 103. The Vault 2018 survey found 52% had participated in an office romance. Id.
  • 114Dana Wilkie, Forbidden Love: Workplace-Romance Policies Now Stricter, SHRM (Sept. 24, 2013), [].
  • 115James Lardner, et al., Cupid’s Cubicles: Office Romance Is Alive and Well, Despite a Barrage of Corporate Countermeasures, U.S. NEWS & WORLD REP., Dec. 14, 1998, at 44, 47.
  • 1169% of women have left work because of failed romance compared to 3% of men. The 2018 Vault Office Romance Survey Results, supra note 110.
  • 117Id.
  • 118Nauen, supra note 103 (finding that 27% of women reporting they have dated at work say they have dated their supervisor compared with 16% of men. The survey suggests that an additional 8% have been higher ranking people not their supervisor. 30% total—35% of women and 25% of men—have dated someone at a higher level in the organization).
  • 119Novak v. Waterfront Comm’n of N.Y. Harbor, 928 F. Supp. 2d 723, 731 (S.D.N.Y 2013).
  • 120Id. at 726.
  • 121Id.
  • 122Id.
  • 123Id.
  • 124Id.
  • 125Id.
  • 126Id. at 726–27.
  • 127Id.
  • 128Id. at 727.
  • 129Id. at 731. The case seems somewhat confused about the elements of a hostile environment claim: “At no point during the relevant period was Novak terminated or suspended, nor did she suffer a loss of pay or other compensation, such as sick time or vacation time. Novak was never demoted or denied an opportunity for promotion, and she was never formally disciplined during her employment at the Commission.” Id. at 727.
  • 130Keppler v. Hinsdale Township High Sch. Dist., 715 F. Supp. 862, 871–72 (N.D. Ill. 1989); see generally Huebschen v. Dep’t of Health and Social Servs., 716 F.2d 1167 (7th Cir. 1983) (Equal Protection clause applied to public employer). To be fair, the Keppler court acknowledged that an explicit quid pro quo (“resume sleeping with me or else”) could violate Title VII, but held that even egregious retaliation based on hurt feelings could not. Keppler, 715 F. Supp. at 870 n.7.
  • 131Cary Franklin, Discriminatory Animus, in A Nation of Widening Opportunities? The Civil Rights Act at Fifty (Samuel Bagenstos & Ellen Katz eds., 2014).
  • 132Oncale, 523 U.S. at 80 (quoting Harris, 510 U.S. at 25 (Ginsburg, J., concurring)).
  • 133See, e.g., Succar v. Dade Cnty Sch. Bd., 229 F.3d 1343, 1345 (11th Cir. 2000) (holding that misconduct does not constitute sexual harassment if based on a “personal feud,” not gender).
  • 134The only exception occurs if the supervisor engages in behavior that would, standing alone and without reference to the past rejection, constitute sexual harassment, such as explicitly conditioning better treatment at work on future romantic or sexual involvement.
  • 135Kelsey Bolar, Why His Female Law Clerks Are among Brett Kavanaugh’s Biggest Advocates, The Daily Signal (July 11, 2018), [];/; Gloria Steinem, Feminists and the Clinton Question, N.Y. Times (Mar. 22, 1998); Callum Borchers, Bill Clinton’s Very Trump-Like Response to Questions about Sexual Misconduct, Washington Post (June 4, 2018), []; USA Today (Jan. 12, 2018), []. (“Three-dozen female former Saturday Night Live employees publicly defended former SNL comic-turned Franken by declaring he never did anything to them.”)
  • 136Fed. R. Evid. 404(b).
  • 137Fed. R. Evid. 404(b)(2).
  • 138Johnson v. United Cerebral Palsy/Spastic Children’s Found., 93 Cal. Rptr. 3d 198, 213 (2009); Soza v. William Ziering, Inc., F035182, 2002 Cal. App. Unpub. LEXIS 6398, at *27–29 (July 11, 2002).
  • 139552 U.S. 379, 388 (2008).
  • 140See, e.g., Bennett v. Nucor Corp., 656 F.3d 802, 809–10 (8th Cir. 2011); Elion v. Jackson, 544 F. Supp. 2d 1, 8 (D.D.C. 2008).
  • 141See, e.g., Johnson v. Interstate Brands Corp., 351 F. App’x, 36, 41 (6th Cir. 2009); Jones v. St. Jude Med. S.C., 823 F. Supp. 2d 699, 734–35 (S.D. Ohio 2011); Lewis v. City of Chicago, 563 F. Supp. 2d 905, 922 (N.D. Ill. 2008); Adams v. Austal, U.S.A., LLC, 754 F.3d 1240, 1258 (11th Cir. 2014).
  • 142See, e.g., Lawson-Brewster v. River Valley Sch. Dist., 617 F. Supp. 2d 589, 595 (W.D. Mich. 2008); Hall v. Mid-State Mach. Prods., 895 F. Supp. 2d 243, 272 (D. Me. 2012); McClendon v. Dougherty, No. 2:10-cv-1339, 2011 WL 4345901, at *4 (W.D. Pa. Sept. 15, 2011).
  • 143Andazola v. Logan’s Roadhouse, Inc., No. CV-10-S-316-NW, 2013 U.S. Dist. LEXIS 73775, at *7 (N.D. Ala. May 24, 2013).
  • 144See, e.g., Elion, 544 F. Supp. 2d at 8; Howard v. D.C. Pub. Sch., 561 F. Supp. 2d. 53, 5553 (D.D.C. 2008); Rodriguez v. Chertoff, No. CIV 05-546-TUC-CKJ, 2008 WL 5087209, at *1 (D. Ariz. Nov. 26, 2008); Pantoja v. Anton, 198 Cal. App. 4th 87, 87 (2011); United Cerebral Palsy/Spastic Children’s Found., 173 Cal. App. 4th at 759; Johnson v. Big Lots Stores, Inc., 253 F.R.D. 381, 386–87 (E.D. La. 2008); Soza, 2002 Cal. App. Unpub. LEXIS 6398.
  • 145See, e.g., Elion, 544 F. Supp. 2d at 8; Howard v. D.C. Pub. Sch., 561 F. Supp. 2d. 53, 55 (D.D.C. 2008); Rodriguez v. Chertoff, No. CIV 05-546-TUC-CKJ, 2008 WL 5087209, at *1 (D. Ariz. Nov. 26, 2008); Johnson v. Big Lots Stores, Inc., 253 F.R.D. 381, 386–87 (E.D. La. 2008). See generally Emma Pelkey, The “Not Me Too” Evidence Doctrine in Employment Law: Courts’ Disparate Treatment of “Me Too” Versus “Not Me Too” Evidence in Employment Discrimination Cases, 92 Or. L. Rev. 545 (2013).
  • 146Rubenfeld, supra note 1, at 1422 n.199 (citing Arthur Ripstein, Force and Freedom: Kant’s Legal and Political Philosophy 15 (2009)).
  • 147See Schulhofer, supra note 1, at ch. 4.
  • 148Id. at ch. 6.
  • 149See generally id.
  • 150The example of J is not Schulhofer’s but is designed to explore his theory.
  • 151Schulhofer, supra note 1, at 186–87.
  • 152Id. at 187.
  • 153Id. at ch. 9.
  • 154Id. at ch. 8.
  • 155Id. at 164–67.
  • 156For example, a number of cases address the issue of preferential treatment of sexual romantic partners. See generally, e.g., Miller v. Dept. of Corrections, 36 Cal. 4th 446 (Cal. 2005); Sherk v. Adesa Atlanta, LLC, 432 F. Supp. 2d 1358 (N.D. Ga. 2006); Stewart v. SBE Entertainment Group, LLC, 239 F. Supp. 3d 1235 (D. Nev. 2017). At some point, the law might well want to concern itself with these, but this extension would be more controversial than those proposed in this article.
  • 157Friedrich Engels, The Origin of the Family, Private Property, and the State (1884).
  • 158Simone de Beauvoir, The Second Sex (H.M. Parshley trans. 1968) (1949).
  • 159Gerda Lerner, The Creation of Patriarchy (1986).
  • 160Shulamith Firestone, The Dialectic of Sex: The Case for Feminist Revolution (1970).
  • 161Baskerville, 50 F.3d at 430.
  • 162John Tooby & Leda Cosmides, Conceptual Foundations of Evolutionary Psychology, in The Handbook of Evolutionary Psychology (David M. Buss ed., John Wiley & Sons 2005). Some evolutionary psychologists regard the emphasis on domain specific mechanisms to be the field’s single most revolutionary contribution to psychology.
  • 163Leda Cosmides & John Tooby, Evolutionary Psychology and the Emotions, in Handbook of Emotions (M. Lewis & J.M. Haviland Jones, eds., 2d ed. 2000).
  • 164See generally Donald Symons, The Evolution of Hum. Sexuality (1981).
  • 165Sarah Blaffer Hrdy, Raising Darwin’s Consciousness: Female Sexuality and the Prehominid Origins of Patriarchy, 8 Hum. Nature 1 (March 1997).
  • 166Jane Lancaster, A Feminist and Evolutionary Biologist Looks and Women, 34 Y.B. of Physical Anthropology 1 (1991).
  • 167Barbara Smuts, The Evolutionary Origins Of Patriarchy, 6 Human Nature 1 (1995); Barbara Smuts, Feminism, the Naturalistic Fallacy, and Evolutionary Biology, 11 Pol. and the Life Sci. 174–76 (1992).
  • 168Robert L. Trivers, Parental Investment and Sexual Selection, in Sexual Selection and the Descent of Man 136–179 (Bernard Campbell ed. 1972).
  • 169Symons, supra note 162.
  • 170Sarah Blaffer Hrdy, “7 Empathy, Polyandry, and the Myth of the Coy Female,” Conceptual Issues in Evolutionary Biology, n.d., 131.
  • 171Lois Romano, The Reliable Source, Washington Post (May 13. 1993), [].
  • 172Gerd Bohner & Norbert Schwarz, The Threat of Rape: Its Psychological Consequences on Non-Victimized Women, in Sex, Power, Conflict: Evolutionary and Feminist Persp. 162–75 (D. Buss and N, Malamuth, eds. 1996).
  • 173This observation, as noted earlier, has a long and distinguished lineage. See, e.g., Hume, supra note 12, at 331–32. More recently it has been made by both the early evolutionary psychologists and their feminist critics. For the early evolutionary psychology perspective, see Symons, supra note162. For their feminist critics, see Lancaster, supra note 164.
  • 174In some societies, women contribute more than men to economic subsistence, and in some the kin of the mother rather than the father play the dominant role in supporting child-rearing. Societies in which paternal support is not important are typically not monogamous, the sexual double standard is relatively relaxed, and women find a diversified portfolio of genetically high-quality partners at least as good as just one. Hrdy, supra note 163; Lancaster, supra note 164. F. W. Marlowe, “The Mating System of Foragers in the Standard Cross-Cultural Sample,” Cross-Cultural Research 37, no. 3 (2003): 282; Marlowe.; Alice Schlegel and H.B. Barry III, “The Cultural Consequences of Female Contribution to Subsistence,” American Anthropologist 88, no. 1 (1986): 142–50.
  • 175Leora Tanenbaum, The Truth about Slut Shaming, Huffington Post (Dec. 6, 2017), [].
  • 176Judge Posner, a complex thinker who on the whole adopts the early sociobiology approach, observed in an article that “solicitations by a supervisor may be resented as signaling the offeror’s unwillingness to recognize that the woman is of high rather than low status.” Richard A. Posner, Employment Discrimination: Age Discrimination and Sexual Harassment, 19 International Review of Law and Economics 421, 437 (1999). This insight seems wholly lost in the Baskerville opinion. Thought the reasons for this are not clear, Posner is generally unsympathetic to the regulation of sexual harassment, and one commentator has suggested that he operates from the assumption that male sexuality is essentially ungovernable, and that women must bear the burden of regulating sex. Id. See also Jane E. Larson, The New Home Economics: A Review of Sex and Reason, 10 Const. Commentary 443, 467 (1993).
  • 177The only place that sex harassment doctrine has even fitfully considered sex differences is the reasonableness standard. See, e.g., Ellison v. Brady, 924 F.2d 872 (9th Cir. 1991).
  • 178Linda Hamilton Krieger, Content of Our Categories: A Cognitive Bias Approach to Discrimination and Equal Employment Opportunity, 47 Stan. L. Rev. 1161, 1244 (1995).
  • 179Krieger and others have gone further, suggesting that the remedies appropriate for unconscious bias are more limited that those appropriate for conscious bias. While I support this position, the analogy between non-misogynistic harassment and unconscious bias do not seem perfect. Full compensatory damages in harassment or any other sex discrimination case should not require proof of misogyny, but there may be some point at which culpability is low enough to justify limitations on damages.
  • 180Alicia Eler, No, Garrison Keillor, Women Who Take Selfies Aren’t “Asking For It”, Harper’s Bazaar (Dec. 4, 2017), [].
  • 181Daphne Merkin, Publicly, We Say #MeToo. Privately, We Have Misgivings., N.Y. Times (Jan. 5, 2018), [].
  • 182Id.
  • 183David Walsh, Ronan Farrow’s Latest #MeToo “Bombshell,” Directed Against Leslie Moonves of CBS, World Socialist Website (Aug. 1, 2018), [].
  • 184Katrin Bennhold, Another Side of #MeToo: Male Managers Fearful of Mentoring Women, N.Y. Times (Jan. 27, 2019), [].